NEWS

Release 3.6.2
=============

Bugs fixed
----------

- :meth:`rpy2.rinterface.initr` was not skipping all initialization steps when the
  embedded R is already initialized by :mod:`rpy2` (contributed to issue #1190).

Changes
-------

- Warnings about R-related environment variables after initialization are downgraded
  to `INFO`-level and emitted by :obj:`rpy2.rinterface.logger` (solves issue #1190).

- On Windows, the determination of where `R.dll` is located now has a fallback if `make` and `bash`
  are not installed (issue #1188).


Release 3.6.1
=============

Bugs fixed
----------

- Clear warnings for deprecation of `rpy2.rinterface_lib.callbacks.obj_in_module`
  (issue #1181).

- Fix callback setup during initialization on Windows (issue #1183).

- In the R magic, setting options such as width, height, point size, etc. was broken
  in the new updated system to manage graphical devices.


Release 3.6.0
=============

New features
------------

- On systems where the default encoding is not utf-8, converting
  R strings to Python can result in decoding errors. The default
  encoding is now the system's encoding (rather than assume utf-8).
  This may provide a fix to issue #1018.

- Noteboook "R magic" features for ipython/jupyter support more
  graphical devices to render R graphics in notebooks. Registered
  devices are in :obj:`rpy2.ipython.rmags.graphics_devices`. This
  allows the specification of fallbacks when the preferred device is
  not available because of a missing R package. For SVG graphics,
  it will first try to use the R package `svglite` to produce
  lighter-weight SVG representations that are better when notebooks
  may include many figures and figure with large number of visual
  elements.

- Experimental minimalist set of optional conversion rules to
  convert Python :class:`list` objects to R vectors when all elements
  are of atomic type: :obj:`rpy2.robjects.flatlist_converter`. Those
  rules have been in used for the ipython/jupyter default converter.

- The setter :attr:`rpy2.robjects.vectors.Array.dim` is now implemented,
  allowing to change the dimension of an R array in place
  (fixes issue #1145). 

Changes
-------

- `rpy2` is now a namespace package, with `rpy2-rinterface`
  (low-level interface close to R's C API) and `rpy2-robjects`
  (possible more-Pythonic high-level interface using the low-level
  interface) 2 packages on pypi installing into that namespace.
  The package `rpy2` remains on pypi as essentially a metapackage
  depending on `rpy2-rinterface` and `rpy2-robjects`.

- :mod:`rpy2.rlike.functional` is deprecated. The functions it
  provides proved of little use and can implemented in one line.
  It will be removed with :mod:`rpy2` 3.7.

- :class:`rpy2.rlike.collections.OrdDict` is deprecated.
  :class:`rpy2.rlike.collections.NamedList` is enough to have
  a Python data structure to represent R lists or paired lists.

- The long-deprecated functions
  :func:`rpy2.robjects.robjects.numpy2ri.activate`,
  :func:`rpy2.robjects.robjects.numpy2ri.deactivate`,
  :func:`rpy2.robjects.robjects.pandas2ri.activate`,
  and :func:`rpy2.robjects.robjects.pandas2ri.deactivate` are no
  longer functional. Calling them will raise an exception pointing
  to documentation.

- The wrapper for ggplot2 is now targeting the R package `ggplot2`
  version 3.5.

- The wrapper for dplyr is now targeting the R package `dplyr`
  version 1.1.
- The default converter for the ipython/jupyter magics depended on
  what is the default converter at the time of module import. To
  simplify this, the default converter is now fixed to
  `robjects.default_converter` + `robjects.flatlist_converter` +
  `robjects.numpy2ri.converter` + `robjects.pandas2ri.converter`.
  If `pandas` cannot be imported
  it falls back to `robjects.default_converter` + `robjects.numpy2ri.converter`.
  If `numpy` cannot be imported it falls back to  `robjects.default_converter`.
- The definition of `Rcomplex` from R's C-API was updated to match
  the definition in recent R release (>= 4.3) when rpy2 is in "API mode".
  This does not work with the "ABI mode" though, and the legacy
  definition of `Rcomplex` is still used with it.
- :func:`rpy2.rinterface.initr_simple` is no longer part of the API. Use
  :func:`rpy2.rinterface.initr` can provide the same functionality.


Bugs fixed
----------

- Fix case of `zoneinfo.ZoneInfo` object missing the attribute `zone`.
  The attribute `key` is used instead. This fixes issue #1079.

- The constructor for :class:`rpy2.robjects.lib.ggplot2` did not
  play well with conversion rules in :mod:`rpy2.robjects.pandas2ri`.
  The default conversion rules are now forced (issue #1116).

- The documentation for the notebook "magic" `Rdevice` was not
  displayed in the module-level docstring for :mod:`rpy2.ipython.rmagic`.


Release 3.5.17
==============

New features
------------

- `geom_curve` added to the `ggplot2` mapping.


Release 3.5.16
==============

Changes
-------

- After the embedded R is initialized the SIGINT handler is set to
  Python default handler (instead of a custom handler raising
  :class:`KeyboardInterrupt`).

Bugs fixed
----------

- :meth:`rpy2.robjects.methods.getclassdef` was not reporting
  clearly errors due unspecified package name or missing class
  definition in within the package namespace specified.

- The display of SVG figures in jupyter notebooks is fixed.
  there was a typo in the code.


Release 3.5.15
==============

New features
------------

- Checking C compiling against the R shared library can be skipped with
  the environment variable `RPY2_API_FORCE` set to `True`.

- The rinterface-level class :class:`rpy2.rinterface.SexpExtPtr` now
  also has an robject-level wrapper :class:`rpy2.robjects.ExternalPointer`,
  and a representation for class name-based conversion rules.


Bugs fixed
----------

- Fixed notebook representation of R named arrays or vectors when
  conversion rules might interfere (issue #1047).

- Multi-line in environment variables (issue #1066).

Changes
-------

- Complete drop of support for Python 3.7. It is now removed from
  the build matrix.


Release 3.5.14
==============

Bugs fixed
----------

- Fixed use of named arguments with `ExtractDelegator` (issue #1048).

- `rpy2py` conversion rules in `numpy2ri` could convert R `list` objects into
  :class:`rpy2.containers.rlike.OrdDict` but there were no `py2rpy` rules
  to perform the reverse conversion (issue #944).

- :meth:`rpy2.robjects.vectors.ListVector.from_length` was incorrectly
  calling the conversion rules before returning the :class:`ListVector`.

Changes
-------

- `distutils` was still used in a couple of places. The installation process
   should now be relying only on `setuptools`.

Release 3.5.13
==============

New features
------------

- R is setting a number of environment variables in a wrapper script. Those
  variables were not set by default by rpy2 and this could lead to issues (see issue #1033).
  The variables are now added to `os.environ`.

Bugs fixed
----------

- `python -m rpy2.situation LD_LIBRARY_PATH` could incorrectly truncate the value
  for `LD_LIBRARY_PATH` when the variable `LD_LIBRARY_PATH` is already defined.  

Changes
-------

- The dependency on :mod:`pytz`, a deprecated package, was
  removed and replaced by a dependency to :mod:`zoneinfo`.
  If Python < 3.9 the package :mod:`backports.zoneinfo`
  is an added dependency.

- The depency on :mod:`tzlocal` does no longer limit it
  to `<5.0` (issue #1035).

- Remove dependency on :mod:`distutils` in Python's standard library. That package is
  deprecated and removed in Python 3.12 (see https://peps.python.org/pep-0632/ and
  issue #1040).


Release 3.5.12
==============

New features
------------

- :func:`rpy2.rinterface.evalr_expr_with_visible` to evaluate
  an expression and return the visibility status of the
  value returned.

- More events in the intialization and termination of an
  embedded R logged by :mod:`rpy2.rinterface_lib.embedded.logger`
  at `INFO` and `DEBUG` levels.

- :meth:`rpy2.robjects.R.__call__` has 2 named arguments "visible" and
  "print_r_warnings" to handle R "invisible" results and print R warnings
  at the end of the evaluation. Invisible results happen when doing
  `rpy2.robjects.r("x <- 1")`. R will return the value of x "invisibly".
  The default are `invisible is True` and `print_r_warnings is True`.


Bugs fixed
----------

- "R magic" cells are now show R warnings at the end of the
  evaluation of an R cell (#issue 226).

Changes
-------

- Evaluating a string as R code using :meth:`rpy2.robjects.R.__call__`
  (e.g., `rpy2.robjects.r("1+2")`) now shows R warnings at the end
  of the evaluation by default.

- :meth:`rpy2.robjects.R.__call__` returns the results invisibly by default
  (see section "New features" for this release).


Release 3.5.11
==============

New features
------------

- :class:`pandas.Categorical` objects are now converted to R
  `factor` objects.

Bugs fixed
----------

- :mod:`pandas` 2.0 makes API-breaking changes that made
  conversion fail (issue #1013).
  
- The default converter for R magic was dropping R factors
  to :mod:`numpy` arrays of integers rather than convert them
  to pandas :class:`Category` (issue #1010).

Changes
-------

- The fix for issue #1010 (see above) changes the numpy conversion
  for R factors. It returns an arrays with the level factors rather
  than the integer codes for the levels.

- :func:`rpy2.rojects.pandas2ri.py2rpy_categoryseries` raises a
  :class:`DeprecationWarning`.


Release 3.5.10
==============

New features
------------

- :mod:`rpy2.situation` reports cffi interface type information.

- The ipython/jupyter R magic can now have `jpeg` as a default graphics format
  for static figures in addition to `png` and `svg`.

Bugs fixed
----------

- Building the path to the R shared library was incorrectly
  using the output of `R CMD config LIBnn` (issue #982).

- R started raising warnings when calling `formals` on `BUILTINSXP` R objects.
  Internal functions in :mod:`rpy.robjects.functions` calling `formals` no
  longer propagate the warnings.


Release 3.5.9
=============

New features
------------

- Type hints in :mod:`rpy2` are now checked with `mypy`.

Bugs fixed
----------

- Building the path to the R shared library was incorrectly
  using the output of `R CMD config LIBnn` (issue #982).

- R started raising warnings when calling `formals` on `SPECIALSXP` R objects.
  Internal functions :mod:`rpy.robjects.functions` calling `formals` no longer
  propagate the warnings.
  
- The :mod:`numpy` converter was not turning `NA_character_` into
  `None` (issue #979).

- :mod:`rpy2.situation` included an eager `import rpy2` that could cause a version
   mismatch error with some build/install toolchain (issue #984). The import is now
   lazy/delayed.

- Installation targets `pandas`, `all`, and `test` now specify `pandas>=1.2.0`
  (which should limit frequencies of issues like #998).

Changes
-------

- :class:`rpy2.robjects.conversion.Converter` objects are no longer functioning context
  managers. An exception is now raised when trying to use it that way.
  The feature was introduced with rpy2-3.5.7 but it does not guarantee the locality
  of a context manager and can result in permanently changed conversion rules.
  The method :meth:`rpy2.robjects.conversion.Converter.context` should be used instead.


Release 3.5.8
=============

New features
------------

- Default value for the interactivity status of the embedded R
  as initialization is now a private variable
  `_DEFAULT_R_INTERACTIVE` in :mod:`rpy2.rinterface_lib.embedded`.

Bugs fixed
----------

- :func:`rpy2.rinterface.initr_checkenv` (aliased to
  :func:`rpy2.rinterface.initr`) did not include
  initialization parameters for `interactive`, `_want_setcallbacks`,
  and `_c_stack_limit` (issue #976).
  
- The numpy converter was not listing R objects of type `CHARSXP` as
  vectors. This make R `NA_character_` values wrapped in `rpy2` Python
  objects to be passed as-is (issue #983).


Release 3.5.7
=============

New features
------------

- :class:`rpy2.robjects.conversion.Converter` objects are now also context
  managers. This makes the use of :func:`rpy2.robjects.conversion.localconverter`
  unnecessary as a converter can now be used directly with a Python `with`
  statement. The documentation was updated accordingly.

Bugs fixed
----------

- Building the C-extension could fail when the C compiler is not
  C99-compatible or requires to be specified C99 compatibility
  (issue #963).

- With the "R magic", converters in a local namespace were inadvertently
  skipped (the local namespace was not searched).

- Calling :func:`rpy2.rinterface.rternalize` before the embedded R is
  initialized was ending with a segfault. It is now raising an
  :class:`rpy2.rinterface_lib/embedded/RNotReady` (issue #971).

Changes
-------

- The default for the :class:`contextvars.ContextVar` holding
  conversion rules (:obj:`rpy2.robjects.conversion.converter`)
  is now a dummy converter raising errors about missing
  conversion rules. This behavior helps with troubleshooting
  Python code using multithreading without fetching the context
  (see issue #965).

- A warning is now issued when trying to convert a `DataFrame` with
  duplicate indexes (issue #811).


Release 3.5.6
=============

Bugs fixed
----------

- :class:`rpy2.rlike.container.OrdDict` objects were triggering an
  error when unpickled (PR #955).

- Function signature mapping not longer assumes that all default
  values are R vectors mapped to a `Vector` instance (PR #954).


Release 3.5.5
=============

New features
------------
  
- For ipython and Jupyter notebooks, R "magics" accepting optional input arguments
  (`-i` or `--input`) can now be of form `<python-name>` or
  `<r-name>=<python-name>`. The former is the way is has been working so far:
  the R object is given the name of the Python object. The latter allows to
  specify the name under which the input object will be known to R. In addition
  to that the python name can optionally be a path, that is a name nested in a
  sequence of namespaces. For example, `-i module.varname`. The magics
  concerned are `%R`, `%%R`, and `%Rpush` (issues #934 and #935).

- For ipython and Jupyter notebooks, R "magics" accepting an optional conversion
  argument (`-c/--converter`) now also accepts names that are a path in Python,
  that is a name nested in a sequence of namespaces. For example,
  `-c rpy2.robjects.default_converter` (issue #934).

- The "magic" `%Rpush` accepts an optional argument `-c/--converter` that is
  similar in nature to the argument of the same name in `%R` and `%%R`.

Bugs fixed
----------

- :class:`rpy2.rlike.container.TaggedList` objects were triggering an error when unpickled (issue #947).

- Converting :class:`pandas.Series` of type `"object"` with a fallback
  to R strings no longer fails with an error (issue #916).

Changes
-------

- :meth:`rpy2.robjects.help.Page.iteritems` is deprecated. Use the method `items()` instead.

- Calling :func:`str` on rpy2 objects that are proxies of R objects is now
  using R's `print()` whenever R's `show()` results in an error (issue #908).


Release 3.5.4
=============

New features
------------

- :func:`rpy2.rinterface.rternalize` can optionally mirror more accurately in R
  the function signature defined in Python (issue #905).

Bugs fixed
----------

- Better handling of timezone in :class:`rpy2.robjects.vectors.POSIXct`
  (issue #917).

- :mod:`pandas` conversion is now correctly handling R data frames with duplicated
  column names. They were dropped tbefore that (issue #925).

- :mod:`pandas` conversion is now correctly handling R data frames with nested
  columns (issue #930).

- Importing :mod:`rpy2.ipython.rmagic` while :mod:`pandas` is not installed resulted
  in an error.


Changes
-------

- :func:`rpy2.robjects.methods.getclassdef` is now explicitly
  naming the optional argument with the R package name when calling
  the underlying R function.

- The `%R` line magic for ipython/jupyter was trying to substitute anything
  after a dolar (`$`) sign, which caused issue since this character is
  used to access by name elements in R containers. No substitution
  is happening anymore (issue #922).


Release 3.5.3
=============

Bugs fixed
----------

- Pandas converter no longer fails with arrays of `dtype`
  :class:`pandas.BoolDtype` or :class:`pandas.Float64Dtype`.(issue #880).

- Numpy and pandas conversion could fail an R object is a low-level wrapper
  for an R list (issue #890).

- Fixed detection of the R install path on Windows (issues #875 and #885).

- Fixed error with dynamic docstring generation for functions
  without arguments (issue #903).

- Fixed error with dynamic docstring generation when R functions
  are missing the section 'Details' (isue #902).


Release 3.5.2
=============

New features
------------

- The string representation for :class:`rpy2.robjects.conversion.Converter`
  objects lists classes the converter will dispatch on.

Bugs fixed
----------

- :meth:`rpy2.robjects.vectors.FactorVector.iter_labels` was failing if
  any element is NA (issue #879).

- The definition of converters was not thread-safe. The fix deprecates
  the use of :obj:`rpy2.robjects.conversion.converter` and replaces it
  with :func:`rpy2.robjects.conversion.get_conversion`.


Release 3.5.1
=============

Bugs fixed
----------

- Fix `__setitem__` on R list when the value is a simple Python object with
  a default low-level conversion (issue #864).

- The default converter for the rmagic could fail with a
  :class:`RecursionError` (issue #866).


Release 3.5.0
=============

New features
------------

- :obj:`rpy2.rinterface.NULL` can be refererenced in code before the embedded
  R is initialized.

- `LD_LIBRARY_PATH` is obtained from the executable `RScript` and set prior
  to initializing R at the :mod:`rpy2.rinterface` level (issue #833).

- The package is now fully typed with type hints.

- The :func:`rpy2.rinterface.evalr` accepts the named parameters `envir`
  and `enclos`, mirroring better the signature of the R function `eval`.

- New function :func:`rpy2.rinterface.evalr_expr` to evaluate R expressions.

- :mod:`rpy2.situation` is now using a module-level (called `logger`) to
  facilitate troubleshooting. When the module is run as an executable
  the flag `--verbose` can take the possible values DEBUG, INFO, WARNING, ERROR.

Changes
-------

- Better customizablity of conversion rules from R objects with `typeof`
  `INTSXP` (array of integers), `VECSEXP` (R list), and `LGLSXP` (array of
  booleans) :mod:`pandas`.

- The representation of R integer is no longer using a comma separator
  for thousands (issue #797).

- Requirement is now Python>=3.7. :mod:`rpy2` is no longer tested with Python 3.6.

- :mod:`rpy2.robject.lib.grid` was refactored to fix the conversion. This
  fixes issue #804. A classes now have a `classmethod` `r` for the R
  constructor (the `__init__` is the Python constructor).

- :obj:`rpy2.robjects.vectors.__all__` is removed.

- Cleaner class design by having :class:`rpy2.rinterface.NumpyArrayMixin`
  removed and replaced by
  :class:`rpy2.rinterface.SexpVectorWithNumpyInterface`.

- With Python < 3.8 the module :mod:`typing_extensions` is now a requirement.

- Initialization and update steps for
  :class:`rpy2.robjects.conversion.NameClassMap` were updated to ensure
  type hints are correct. This probably solved cryptic bugs with conversion
  system.

- :meth:`rpy2.rinterface.LangSexpVector.from_string` was at the robject level (class method of
  :class:`rpy2.robjects.language.LangVector`) but is now at the interface level.

- :meth:`rpy2.rinterface.get_evaluation_context` is deprecated. The module
  level context variable `evaluation_context` should be used instead. The
  use of the context variable should address issues when using nested
  :meth:`rpy2.rinterface.local_context`.

- Added recommendation to try :meth:`pandas.DataFrame.infer_objects` when
  the converion to R data frames fails because of mixed types (issue #794).

- No longer implicit dependency on :mod:`numpy` for memory views. The Python
  API has an unresolved issue around handling FORTRAN-ordered arrays without
  going to C-level (https://bugs.python.org/issue34778), and C-extension
  as added to rpy2 to cut the dependency on numpy for just this (PR #856).
   
Bugs fixed
----------

- R warning about partial argument name matching is now cleared (issue #821).

- Setting a slice for a StrSexpVector where some of the elements are not strings
  no longer fails.

- Assigning items to an R array of strings using a slice of more than one
  element was likely to fail.

- The :mod:`rpy2.rinterface`-level conversion of Python integers to R integers is now
  checking that integers are not larger than what R can handle.
  
- Trying to access items in an R environment before R is initialized is no
  longer ending with a segfault (issue #842).


Release 3.4.6
=============

- Added a class and an alias to map ggplot2's `element_line` (issue #822).


Release 3.4.5
=============

Changes
-------

- The deprecation warning when using :func:`rpy2.robjects.lib.grid.activate`
  was missing (indirectly revealed through issue #804).

- The named argument `LINPACK` in :meth:`rpy2.robjects.vectors.Matrix.svd`
  is no longer present in R.

Bugs fixed
----------

- SIGPIPE sent to a process running Python+rpy2 could result in a segfault.
  This was caused by an incorrect setting of R signal handlers (issue #809).


Release 3.4.4
==============

Changes
-------

- `RRuntimeError` exceptions raised while evaluating R code
  an R magic (ipython/jupyter) are now propagated (issue #792).
  


Release 3.4.3
=============

New features
------------

- :mod:`rpy2.robjects.lib.ggplot2` maps more functions in the
  R package (issue #767)

- Utility function :func:`rpy2.robjects.lib.ggplot2.dict2rvec`
  to convert a Python `Dict[str, str]` into an R named vector
  of strings.

Bugs fixed
----------

- Calling mod:`rpy2.situation` to report on the environment no longer
  stops with an uncaught exception when no R home can be determined
  (issue #774)

- Converting pandas series with the older numpy types could result
  in an error (issue #781)

- Numpy converter was not properly turing R integer or float arrays
  into their numpy equivalent (issue #785)

- The HTML representation of R list without named element was
  incorrect (issue #787)

- Fix for when R is installed in a user home directory on Windows (PR #788)


Release 3.4.2
=============

Bugs fixed
----------

- Multithreading during the initialization of the embedded R no longer
  triggers a fatal error (issue #729)

Changes
-------

- :mod:`pytest` is now an optional package. Optional sets of packages are
  `numpy`, `pandas`, `test`, and `all` (all optional packages). They
  can be specified during the installation. For example
  `pip install rpy2[test]`. (issue #670)


Release 3.4.1
=============

Bugs fixed
----------

- The file `requirements.txt` was missing from the source distribution
  on pypi (issue #764).


Release 3.4.0
=============

New Features
------------

- The mapping of the R C API now includes `Rf_isSymbol()`.

- Singleton class :class:`rpy2.rinterface_lib.sexp.RVersion` to report
  the R version for the embedded R.

- :func:`rpy2.rinterface.local_context` to create a context manager
  to evaluate R code within a local environment.

- The `staticmethod` :meth:`rpy2.robjects.vectors.DateVector.isrinstance`
  will tell whether an R objects is an R `Date` array.

Changes
-------

- The dynamic generation of docstrings for R man pages
  is now using R's `Rd2txt`.

- The :func:`rpy2.rinterface_lib._rinterface_capi._findVarInFrame`
  is replaced by the function
  :func:`rpy2.rinterface_lib._rinterface_capi._findvar_in_frame`
  (see fix to issue #710).

- The functions :func:`rpy2.robjects.numpy.activate()` and
  :func:`rpy2.robjects.pandas.activate()` are deprecated and will
  be removed in rpy2-3.5.0.

- :func:`rpy2.rinterface_lib.embedded.setinitialized` was renamed to
  :func:`rpy2.rinterface_lib.embedded._setinitialized` to indicate that
  one should not use it.

- :meth:`rpy2.robjects.lib.ggplot2.vars` to map the R function
  `ggplot2::vars` (issue #742).

- Report correctly the class of R matrix objects with R>=4.0: it is
  now `('matrix', 'array')`. With R<4.0 `('matrix')` is still reported.

- The conversion of R/rpy2 objects to python objects using R class name mapping
  is extended to more classes. The documentation about conversion covers the topic.

- If `R_NilValue` is not null when the initialization of the embedded R is attempted,
  it is now assumed that R was initialized through other means (e.g., an other C library in the
  same process) and the C-level initialization is be skipped.

- The conversion `rpy2py` is now working with any Python object inheriting
  from `_rinterface_capi.SupportsSEXP`.

Bugs fixed
----------

- The C function `Rf_findVarInFrame()` in the R API can trigger
  in an R-level error, and while this is rare, when it does
  when embedded in Python it creates a segfault. Calls are
  now wrapped in `R_ToplevelExec()` to limit the propagation
  of R exceptions. This solved issue #710.

- More complete and correct mapping of R class names in
  :func:`rpy2.rinterface_lib.sexp.rclass_get`.
  
- Initializing the embedded R caused the loss of ability to use Ctrl-C
  to send SIGINT to a Python process (issue #723)

- :mod:`rpy2.sitation` is now working when the environment variable
  `R_HOME` is set even though R is not in the `PATH` or in the Windows
  registry (issue #744).

- Handling an R language objects could result in a segfault when its
  R class was queried (issue #749).

- The conversion of R string arrays to `numpy` arrays was leaving
  R's `NA` value as R NA objects. NAs in this type of arrays are now
  turned to `None` in the resulting `numpy` array (issue #751).

- `rpy2.situation.get_rlib_path()` was returning an environment variable
  with an invalid separator on Windows (mentioned in issue #754).

- R strings encoded with something else than 'utf-8' could result in
  errors when trying to convert to Python strings (issue #754).

- Extracting documentation pages for R objects in packages could
  generate spurious warnings when several "section" tags are present.

- R `Date` arrays/vectors were not wrapped into
  :class:`rpy2.robjects.vectors.DateVector` objects but left as
  R arrays of floats (which they are at the C level).

- The HTML representation of short R lists without names could
  fail with an error.

- The :meth:`__repr__` of `robjects`-level objects was not displaying
  the rpy2 class the R object is mapped to.


Release 3.3.6
=============

Bugs fixed
----------

- The unit tests for importing R packages with `lib_loc` were
  broken (issue #720).

- Trying to create a memoryview for an R array with complex values
  was failing with an attribute error.

- Fix the constructor of metaclass
  :class:`rpy2.robjects.methods.RS4Auto_Type`.

- Fix call to end the embedded R in :class:`rpy2.robjects.R.__cleanup__`
  (issue #734).


Release 3.3.5
=============

Bugs fixed
----------

- The callback handler to read input to R returned an
  invalid result, leading to R asking for input
  without ever acknowledging it received it.


Release 3.3.4
=============

Bugs fixed
----------

- Creating an R vector object from a Python object implementing
  the buffer protocol could give incorrect results as C-level
  incompatibilities could be missed (issue #702).

- :func:`rpy2.robjects.packages.importr` could fail when `lib_loc`
  was specified (issue #705).


Release 3.3.3
=============

Bugs fixed
----------

- Fallback for when `str2lang` is missing (R < 3.6)

- Fix segfault with :meth:`PairListSexpVector.__getitem__` when
  elements of the R pairlist have a `NILSXP` name (issue #700)


Release 3.3.2
=============

Bugs fixed
----------

- Initial fixes to have rpy2 running in ABI mode on Windows.
  Few tests are not passing (many in callbacks for R's C API).

- System detection is now checking for FreeBSD.


Release 3.3.1
=============

Bugs fixed
-----------

- :meth:`rpy2.robjects.conversion.NameClassMap.update` can update
  the mapping (:class:`dict`) or the default class.

Changes
-------

- Adding local converters was overwriting the base `NameClassMap`.


Release 3.3.0
=============

New features
------------

- Trying to import an R package that is not installed will now raise an
  exception :class:`rpy2.robjects.packages.PackageNotInstalledError`.

- The R C API functions `void SET_FRAME(SEXP x, SEXP v)`,
  `void SET_ENCLOS(SEXP x, SEXP v)` and `void SET_HASHTAB(SEXP x, SEXP v)`
  are now accessible through rpy2.

- The module :mod:`rpy2.situation` can now return `LD_LIBRARY_PATH`
  information about R. For example with
  `python -m rpy2.situation LD_LIBRARY_PATH`

- :meth:`rpy2.robjects.methods.RS4.extends` lists the class names in the
  inheritance line.
  
- The conversion of R objects to Python allows much more flexibility
  and better allow the use of independent code converting different classes.
  This is currently limited to R objects that are lists, environments, or
  S4 objects. The Sphinx documentation contains an example. While this is
  still work in progress this should already address concerns
  at the origin of issue #539 about S4 classes.

- :class:`rpy2.robjects.language.LangVector` to map R language objects at
  the `robjects` level.

- :class:`rpy2.robjects.vectors.PairlistVector` to map R pairlist objects at
  the `robjects` level.

- An alternative function to display the output of R cells can be
  specified using `-d` or `--display` in the magic arguments
  (in :mod:`rpy2.ipython.rmagic`).

- Python classes representing underlying R objects no longer have to
  exclusively rely on inheritance from :mod:`rpy2.rinterface` objects`.
  An abstract class :class:`rpy2.rinterface_lib.sexp.SupportsSEXP` is added
  to identify objects supporting a `__sexp__` protocol, and that abstract
  class can also be used with type hints.
- :func:`rpy2.robjects.functions.wrap_r_functions` can create Python functions
  with matching signature from R functions

- :func:`rpy2.robjects.functions.wrap_r_functions` can create Python functions
  with matching signature from R functions.

- New class :class:`rpy2.rinterface_lib._rinterface_capi.UninitializedRCapsule`
  to allow the instanciation of "placeholder" rpy2 objects before the
  embedded R is initialized. This facilitate the use of static typing checks
  such as mypy, mocking for tests that do not involve the execution of R
  code, and allow cleaner implementations of module-level globals
  that are R objects.

- New class :class:`rpy2.robjects.vectors.DateVector` to represent R dates.

- :class:`pandas.Series` containing date objects can now be converted to R
  `Date` vectors.

Changes
-------

- When calling R C-API's `R_ParseVector` and a error occurs, the
  exception message now contains the parsing status.

- :mod:`rpy2.rinterface_lib.embedded` has a module-level "constant"
  `DEFAULT_C_STACK_LIMIT` used when initializing the embedded R.

- When creating a :mod:`rpy2.robjects.vectors.DataFrame` from (name, vector)
  pairs, the names are no longer transformed to syntactically valid R
  symbols (issue #660).

- The value `nan` in :mod:`pandas` Series with strings is now converted
  to R NA (issue #668).

- Initial support for :const:`pandas.NA` (still experimental in pandas
  at the time of writing, and rpy2 support is limited to arrays of strings).

- :mod:`pandas` series of dtype :class:`pandas.StringDType`, experimental in pandas 1.0,
  are now supported by the converted (in the pandas-to-R direction) (issue #669)

- Version checking for the mapping of R packages in :mod:`rpy2.robjects.lib` is
  now more permissive (check that version prefixes are matching).

Bugs fixed
-----------

- Building ABI only mode could require an API build environment (and fail
  with an error when not present).
  
- SVG output for the R magic were incorrectly bytes objects.

- :meth:`rpy2.rinterface_lib.sexp.StrSexpVector.__getitem__` was returning the string
  `'NA'` when an R NA value. Not it returns `rpy2.rinterface_lib.na_values.NA_Character`.


Release 3.2.7
=============

Bugs fixed
----------

- An f-string in `_rinterface_cffi_build.py` prevented installation
  on Python 3.5 (issue #654).


Release 3.2.6
=============

Bugs fixed
----------

- The conversion of date/time object with specified timezones
  was wrong when different than the local time zone (issue #634)

- Iterating over :mod:`rpy2.situation.iter_info()` could result
  in a error because of a typo in the code.

Changes
-------

- :mod:`pandas` 1.0.0 breaks the conversion layer. A warning
  is now emitted whenever trying to use `pandas` >= 1.0.


Release 3.2.5
=============

Bugs fixed
----------

- Latest release for R package `rlang` broke import through `importr()`.
  A workaround for :mod:`rpy2.robjects.lib.ggplot2` is to rename the
  offending R object (issue #631).

Changes
-------

- f-string requiring Python >= 3.6 removed.


Release 3.2.4
=============

Bugs fixed
----------

- An incomplete backport of the bug fixed in 3.2.3 broke the ABI mode.


Release 3.2.3
=============

Bugs fixed
-----------

- Error when parsing strings as R codes could result in a segfault. 


Release 3.2.2
=============

Bugs fixed
----------

- Python format error when trying to report that the system is not reported
  on Windows (issue #597).

- The setup script would error on build if R is not installed. It is now
  printing an error message.


Release 3.2.1
=============

Bugs fixed
----------

- The wrapper for the R package `dbplyr` could not import the underlying
  package (refactoring elsewhere was not propagated there).

- Creating R objects called `names` `globalenv` caused the method
  :meth:`Sexp.names` to fail (issue #587).

- Whenever the pandas conversion was activated :class:`FloatSexpVector` instances
  with the R class `POSIXct` attached where not corrected mapped back to pandas
  datetime arrays. (issue #594).

- Fix installation when an installation when a prefix without write access is used
  (issue #588).


Release 3.2.0
=============

New features
------------

- rpy2 can built and used with :mod:`cffi`'s ABI or API modes (releases 3.0.x and
  3.1.x were using the ABI mode exclusively). At the time of writing the default
  is still the ABI mode but the choice can be controlled through the environment variable
  `RPY2_CFFI_MODE`. If set, possible values are `ABI` (default if the environment
  variable is not set), `API`, or `BOTH`. When the latter, both `API` and `ABI`
  modes are built, and the choice of which one to use can be made at run time.

Changes
-------

- The "consoleread" callback (reading input to the R console) is now assuming UTF-8
  (was previously assuming ASCII) and is no longer trying to add a "new line" character
  at the end of the input.

- Querying an R environment with an invalid key will generate a :class:`TypeError`
  or a :class:`ValueError` depending on the issue (rather than always :class:`ValueError`
  before.

Bugs fixed
----------

- `setup.py` is now again compatible with Python2 (issue #580).

- Unit tests were failing if numpy is not installed.

- :mod:`rpy2.situation` is no longer breaking when R is not the in path and
  there is no environment variable `R_HOME`.

- Build script for the cffi interface is now using the environment
  variable `R_HOME` whenever defined (rather that always infer it from the
  R in the PATH).

- Converting R strings back to Python was incorrectly using `Latin1` while `UTF-8` was
  intended (issue #537).


Release 3.1.0
=============

New features
------------

- Python matrix multiplication (`__matmul__` / `@`) added to
  R :class:`Matrix` objects.

- An :class:`threading.RLock` is added to :mod:`rpy2.rinterface_lib.openrlib` and is
  used by the context manager :func:`rpy2.rinterface_lib.memorymanagement.rmemory`
  to ensure that protect/unprotect cycles cannot be broken by thread switching, at least
  as long as the context manager is used to handle such cycles (see issue #571).

- The documentation covers the use of notebooks (mainly Jupyter/Jupyterlab).

- The PNG output in Jupyter notebooks R cells can now specify an argument `--type`
  (passed as the named argument `type` in the R function `png`).
  For example on some Linux systems and R installations, the type `cairo`
  can fix issues when alpha transparency is used.

Changes
-------

- Added callbacks for `ptr_R_Busy()` and `ptr_R_ProcessEvents()`.

- `rstart` now an objects in :mod:`rpy2.rinterface_lib.embedded`
  (set to `None` until R is initialized).

- Unit tests are included in a subpackage :mod:`rpy2.tests` as was the
  case before release 3.0.0 (issue #528).

- Experimental initialization for Microsoft Windows.

- :mod:`rpy2.situation` is now also reporting the rpy2 version.

- :func:`rpy2.robjecs.package_utils.default_symbol_check_after` was
  renamed  :func:`rpy2.robjecs.package_utils.default_symbol_resolve`.
  The named parameters `default_symbol_check_after` present in few methods
  in :mod:`rpy2.robjects.packages` and :mod:`rpy2.robjects.functions` were
  modified to keep a consistent naming.

- Trying to instantiate an :class:`rpy2.rlike.container.OrdDict` with a
  a :class:`dict` will result in a :class:`TypeError` rather than a
  :class:`ValueError`.

- Methods of :class:`rpy2.rlike.container.OrdDict` now raises a
  :class:`NotImplementedError` when not implemented.

- The creation of R vectors from Python sequences is now relying on a method
  :meth:`_populate_r_vector` that allows vectorized implementation to
  to improve speed.

- Continuous integration tests run against Python 3.6, 3.7, and 3.8. It is
  no longer checked against Python 3.5.

Bugs fixed
----------

- `aes` in :mod:`rpy2.robjects.lib.ggplot2` had stopped working with the
  R package ggplot2 reaching version 3.2.0. (issue #562).

- Better handling of recent :mod:`pandas` arrays with missing values
  (related to issue #544).

- The mapping of the R operator `%in%` reachable through the attribute `ro`
  of R vectors was always returning `True`. It is now working properly.

- R POSIXct vectors with `NA` dates were triggering an error when converted
  in a data frame converted to :mod:`pandas` (issue #561).


Release 3.0.5
=============

Bugs fixed
----------

- No longer allow installation if Python 3 but < 3.5.

- Fixed error `undefined symbol: DATAPTR` if R < 3.5 (issue #565).


Release 3.0.4
=============

Bugs fixed
----------

- Fixed conversion of `pandas` :class:`Series` of dtype `pandas.Int32Dtype`,
  or `pandas.Int64Dtype` (issue #544).


Release 3.0.3
=============

Bugs fixed
----------

- Fixed the evaluation of R code using the "R magic" was delaying all
  output to the end of the execution of that code, independently of
  whether the attribute `cache_display_data` was `True` or `False`
  (issue #543).

- Fixed conversion of :class:`pandas.Series` of `dtype` "object" when
  all items are either all of the same type or are :obj:`None` (issue #540).


Release 3.0.2
=============

Bugs fixed
----------

- Failing to import `pandas` or `numpy` when loading the "R magic" extension
  for jupyter/ipython was hiding the cause of the error in the `ImportError`
  exception.

- Fallback when an R `POSIXct` vector does not had an attribute `"tzone"`
  (issue #533).

- Callback for console reset was not set during R initialization.

- Fixed rternalized function returning rpy2 objects (issue #538).

- `--vanilla` is no longer among the default options used to initialize R
  (issue #534).


Release 3.0.1
=============

Bugs fixed
----------

- Script to install R packages for docker image never made it to version
  control.

- Conversion of R arrays/matrices into numpy object trigged a segfault
  during garbage collection (issue #524).


Release 3.0.0
=============

New features
------------

- rpy2 can be installed without a development environment.

- Unit tests are now relying on the Python module `pytest`.

- :attr:`rpy2.rinterface.NA_Integer` is now only defined when the embedded R
  is initialized.

Changes
-------

- complete rewrite of :mod:`rpy2.rinterface`.
  :mod:`cffi` is now used to interface with the R compiled shared library.
  This allows ABI calls and removes the need to compile binaries. However, if
  compilation is available (when installing or preparing pre-compiled binaries)
  faster implementations of performance bottlenecks will be available.
  
- calling :func:`rpy2.rinterface.endr` multiple times is now only ending R
  the first time it is called (note: an ended R cannot successfully be
  re-initialized).

- The conversion system in the mod:`rpy2.robjects.conversion` now has only
  two conversions `py2rpy` and rpy2py`. `py2rpy` tries to convert any
  Python object into an object rpy2 can use with R and `rpy2py` tries
  to convert any rpy2 object into a either a non-rpy2 Python object or
  a mod:`rpy2.robjects` level object.

- The method `get` for R environments is now called `find()` to avoid
  confusion with the method of the same name in Python (:meth:`dict.get`).

- :class:`rpy2.robjects.vectors.Vector`, :class:`rpy2.robjects.vectors.Matrix`,
  and :class:`rpy2.robjects.vectors.Array` can no longer be used to create
  R arrays of unspecified type. New type-specific classes (for example for
  vectors :class:`rpy2.robjects.vectors.IntVector`,
  :class:`rpy2.robjects.vectors.BoolVector`,
  :class:`rpy2.robjects.vectors.FloatVector`,
  :class:`rpy2.robjects.vectors.ComplexVector`, or
  :class:`rpy2.robjects.vectors.StrVector`) should be used instead.

- mod:`rpy2.rpy_classic`, an implementation of the `rpy` interface using
  :mod:`rpy2.rinterface` is no longer available.

- :class:`rpy2.robjects.ParsedCode` and
  :class:`rpy2.robjects.SourceCode` are moved to
  :class:`rpy2.robjects.packages.ParsedCode` and
  :class:`rpy2.robjects.packages.SourceCode`.

Bugs fixed
----------

- Row names in R data frames were lost when converting to pandas data frames
  (issue #484).

Known issues
------------

- Mismatch between R's POSIXlt `wday` and Python time struct_time's `tm_wday`
  (issue #523).


Release 2.9.6
=============

Bugs fixed
----------

- Latest release of :mod:`pandas` deprecated :meth:`DataFrame.from_items`.
  (issue #514).

- Latest release of :mod:`pandas` requires categories to be a list
  (not an other sequence).

Known issues
------------

- The numpy buffer implemented by R arrays is broken for complex numbers


Release 2.9.5
=============

Bugs fixed
----------

- Missing values in pandas :class:`Category` series were creating
  invalid R factors when converted (issue #493).


Release 2.9.4
=============

Bugs fixed
----------

- Fallback for failure to import numpy or pandas is now dissociated from
  failure to import :mod:`numpy2ri` or :mod:`pandas2ri` (issue #463).

- :func:`repr` for R POSIX date/time vectors is now showing a string
  representation of the date/time rather than the timestamp as a float
  (issue #467).

- The HTML representation of R data frame (the default representation in the
  Jupyter notebook) was displaying an inconsistent number of rows
  (found while workin on issue #466).

- Handle time zones in timezones in Pandas when converting to R data frames
  (issue #454).

- When exiting the Python process, the R cleanup is now explicitly request
  to happen before Python's exit. This is preventing possible segfaults
  the process is terminating (issue #471).

- dplyr method `ungroup()` was missing from
  :class:`rpy2.robjects.lib.dplyr.DataFrame` (issue #473).


Release 2.9.3
=============

Bugs fixed
----------

- Delegate finding where is local time zone file to either a user-specified
  module-level variable `default_timezone` or to the third-party
  module :mod:`tzlocal` (issue #448).


Release 2.9.2
=============

Changes
-------

- The pandas converter is converting :class:`pandas.Series` of `dtype` `"O"`
  to :class:`rpy2.robjects.vectors.StrVector` objects, issueing a warning
  about it (See issue #421).

- The conversion of pandas data frame is now working with columns rather
  than rows (introduce in bug fix for issue #442 below) and this is expected
  to result in more efficient conversions.

Bugs fixed
----------

- Allow floats in figure sizes for R magic (Pull request #63)

- Fixed pickling unpickling of robjects-level instances,
  regression introduced in fix for issue #432 with release 2.9.1 (issue #443).

- Fixed broken unit test for columns of `dtype` `"O"` in `pandas` data frames.

- Fixed incorrect conversion of R factors in data frames to columns of
  integers in pandas data frame (issue #442).


Release 2.9.1
=============

Changes
-------

- Fixing issue #432 (see Section Bugs fixed below) involved removed the method
  `__reduce__` previously provided for all rpy2 objects representing R objects.

Bugs fixed
----------

- An error when installing with an unsupported R version was fixed (issue #420).

- The docstring for `rinterface.endr()` was improperly stating that the function was not taking
  any argument (issue #423).

- Target version of dplyr and tidyr are now 0.7.4 and 0.7.2 respectively. 

- Fixed memory leak when pickling objects (issue #432). Fixing the leak caused a
  slight change in the API (see Section Changes above).

- Conversion to :mod:`pandas` now handling R ordered factor (issue #398).

- :mod:`jinja2` was not listed as a dependency (issue #437).


Release 2.9.0
=============

New features
------------

- New module :mod:`rpy2.situation` to extract and report informations
  about the environment, such as where is the R HOME, what is the
  version of R, what is the version of R rpy2 was built with, etc...
  The module is also designed to be run directly and provide diagnostics:
  `python -m rpy2.situation`.

- :meth:`Environment.values`, :meth:`Environment.pop`,
  :meth:`Environment.popitems`, :meth:`Environment.clear`
  to match :meth:`dict.values`,
  :meth:`dict.pop`, :meth:`dict.popitems`, :meth:`dict.clear`.

- :class:`VectorOperationsDelegator` now has a method `__matmul__` to implement
  Python's matrix multiplication operator (PEP-0645).

- A rule to convert R POSIXct vectors to pandas Timestamp vectors was added (issue #418).

- method :meth:`_repr_html_` for R vectors to display HTML in jupyter.


Changes
-------

- Starting several times the singleton :class:`EventProcessor` longer results
  in a :class:`RuntimeError`. This is now only a warning, addressing
  issue #182.

- The target version for the R package `dplyr` mapped is now 0.7.1, and
  :func:`rpy2.robjects.lib.dplyr.src_dt` (issue #357) and
  :func:`rpy2.robjects.lib.dplyr.src_desc` are no longer present.

- :meth:`Environment.keys` is now a iterator to match :meth:`dict.keys`,
  also an interator in Python 3.

- Target version of `ggplot2` library is 2.2.1.

- Option `stringsasfactors` in the constructor for the class `DataFrame`. If `False`, the
  strings are no longer converted to factors. When converting from pandas data frames
  the default is to no longer convert columns of strings to factors.

- The R "magic" for jupyter is now more consistently using the conversion system, and the
  use of custom converters through the magic argument `-c` will work as expected.

- Docker-related files moved to directory docker/ (where variants image for rpy2 are available)


Bugs fixed
----------

- :func:`numpy.float128` is not available on all platforms. The unit test
  for it is now skipped on systems where it is not present (issue #347)

- R pairlist objects can now be sliced (and issue #380 is resolved).

- Passing parameters names that are empty string to R function was
  causing a segfault (issue #409).

- Trying to build an atomic R vector from a Python object that has a length,
  but it not a sequence nor an iterator was causing a segfault (issue #407).


Release 2.8.6
=============

Bugs fixed
----------

- Trying to build an atomic R vector from a Python object that has a length,
  but it not a sequence nor an iterator was causing a segfault (issue #407 -
  backport from rpy2-2.9.0).


Release 2.8.5
=============

Bugs fixed
----------

- The defintion of the method :class:`rpy2.rlike.container.OrdDict.items`
  was incorrect, and so was the documentation for `rcall` (issue #383)
  
- Giving an empty sequence to :meth:`robjects.sequence_to_vector` is now
  raising a :class:`ValueError` rather than fail with an
  :class:`UnboundLocalError` (see issue #388).

- :meth:`robjects.robject.RSlots.items` is now working (see pull request #57).


Release 2.8.4
=============

Bugs fixed
----------

- The context manager :func:`rpy2.robjects.lib.grdevices.render_to_file`
  is no longer trying to impose a file name generated by :mod:`tempfile`
  (issue #371)

- The symbol `LISTSXP` (corresponding to R pairlist objects) was not
  imported from the C module in rpy2.

- The functions `scale_linetype_discrete` and `scale_linetype_continuous`
  in ggplot2 were not wrapped by :mod:`rpy2.robjects.lib.ggplot2`
  (issue #381)


Release 2.8.3
=============

Bugs fixed
----------

- Fixed the error when the transformation of R "man" pages into Python
  docstrings was failing when the section "arguments" was missing
  (issue #368)
  
- Failing to find R in the PATH during the installation of rpy2
  is now printing an error message instead of a warning (issue #366)
  
  
Release 2.8.2
=============

Bugs fixed
----------

- R's `dplyr::src_dt` was moved to `dtdplyr::src_dt` with `dplyr` release 0.5.0.
  To address this, `src_dt` will become a `None` if the R package `dplyr` is
  discovered to be of version >= 0.5.0 at runtime. (issue #357)

- Conversion issue when R symbols were accessed as attribute of the singleton
  :class:`rpy2.robjects.R`. (issue #334)

- The `rmagic` extension for `ipython` was no longer loading with the latest
  ipython (version 5.0.0). (issue #359)
  
Changes
-------

- The fix to issue #357 (see bugs fixed above) was expanded to cover all
  R packages wrapped in :mod:`rpy2.robjects.lib` and ensure that the respective
  Python modules can loaded even if symbols are no longer defined in future
  versions of the corresponding R packages.


Release 2.8.1
=============

New features
------------

- `Dockerfile` with automated build on dockerhub (https://hub.docker.com/r/rpy2/rpy2)

Bugs Fixed
----------

- Trying to install rpy2 while R is not in the `PATH` resulted in an error
  in `setup.py`.


Release 2.8.0
=============

New features
------------

- New class :class:`rpy2.robjects.SourceCode`. The class extends Python's
  :class:`str` and is meant to represent R source code. An HTML
  renderer for the ipython notebook (syntax highlighting using
  :mod:`pygment` is also added).
  
- New module :mod:`rpy2.robjects.lib.tidyr` providing a custom
  wrapper for the R library `tidyr`

- The long-deprecated functions :func:`rpy2.rinterface.set_writeconsole` and
  :func:`rpy2.rinterface.get_writeconsole` are no longer available. One of
  :func:`rpy2.rinterface.set_writeconsole_regular` / :func:`rpy2.rinterface.set_writeconsole_warnerror`
  or :func:`rpy2.rinterface.get_writeconsole_regular` / :func:`rpy2.rinterface.get_writeconsole_warnerror`
  respectively should be used instead.
  
- The attribute :attr:`rpy2.robjects.RObject.slots` can now be implictly interated on
  (the method :meth:`__iter__` is now an alias for :meth:`keys`).

- The default Python-R conversion is now handling functions. This means that
  Python function can directly be used as parameters to R functions (when
  relevant).

- Ipython display hook `display_png` for ggplot2 graphics.

- :mod:`pandas` "category" vectors are better handled by the pandas conversion.

- New module :mod:`rpy2.robjects.lib.grdevices` providing a custom
  wrapper for the R library 'grDevices', exposing few key functions in
  the package and providing context managers (`render_to_file` and
  `render_to_bytesio`) designed to simplify the handling of static plots
  (e.g., webserver producing graphics on the fly or figure embedded in a
  Jupyter notebook).

- Numpy conversion is handling better arrays with `dtype` equal to `"O"`
  when all objects are either all inheriting from :class:`str` or from
  :class:`bytes`. Such arrays are now producing :class:`StrSexpVector` or
  :class:`BytesSexpVector` objects respectively.
  
- R's own printing of warnings if now transformed to warnings of type
  `rinterface.RRuntimeWarning` (it used to be a regular `UserWarning`)

- The family of functions `src_*` and the function `tbl` in the R package
  `dplyr` have aliases in the module :mod:`rpy2.robjects.lib.dplyr`, and
  a class :class:`DataSource` has been added for convenience.

- :class:`rpy2.robjects.vectors.DataFrame` has a method `head` corresponding
  to R's method of the same name. The method takes the n first row of a
  data frame.

- dplyr's functions `count_` and `tally` are now exposed as methods for
  the class :class:`dplyr.DataFrame`.

Changes
-------

- Building/installing rpy2 with a development version of R does not require
  the use of option `--ignore-check-rversion` any longer. A warning is
  simply issue when the R version is "development".

- On MSWindows, the dependency on `pywin32` was removed (issue #315)

- :class:`GroupedDataFrame` in the dplyr interface module is now inheriting
  from the definition of DataFrame in that same module (it was previously
  inheriting from :class:`robjects.vectors.DataFrame`).

- The default `repr()` for R objects is now printing the R classes
  (as suggested in issue #349).

Bugs Fixed
----------

- Parameter names to R function that are in UTF-8 are no longer causing a
  segfault (issue #332)
  
- Looking for a missing key in an R environment (using `__getitem__` or `[`)
  could raise a `LookupError` instead of a `KeyError`.

- R environment can now handle unicode keys as UTF-8 (was previously
  trying Latin1)

- rpy2 is interrupting attempts to install with Python < 2.7 with an
  informative error message (issue #338)

- Setting the R class can be done by using a simple Python string (issue #341)

- `rpy2.robjects.lib.grid.viewport` is now returning an instance of class
  `Viewport` (defined in the same module) (issue #350)
  

Release 2.7.9
=============

Bug fixed
---------

- Python objects exposed to R could lead to segfault when the Python process is
  exiting (issue #331)


Release 2.7.8
=============

Bugs fixed
----------

- American English spelling was missing for some of the function names
  to specify colour (color) scales.
  
- Fix for printing R objects on Windows (pull request #47)


Release 2.7.7
=============

Bugs fixed
----------

- Pickling `robjects`-level objects resulted in `rinterface`-level objects
  when unpickled (issue #324).


Release 2.7.6
=============
 
Changes
-------

- :mod:`rpy2.robjects.lib.ggplot2` was modified to match the newly released
  ggplot2-2.0.0. This is introducing API-breaking changes, which breaks the
  promise to keep the API stable through bugfix releases within series, but
  without it 2.7.x will not a work with new default installation of the R
  package ggplot2.


Release 2.7.5
=============

Bugs fixed
----------

- Division and floordivision through the delegator `.ro` provided with
  R vectors wrapped by `robjects`. (issue #320)

- Memory leak when unserializing (unpickling) R objects bundled in Python
  objects (issue #321)
  

Release 2.7.4
=============

Bugs fixed
----------

- Python 3.5 highlighted slightly incorrect C-level flags in rpy2 objects
  declarations, and :mod:`rpy2.robjects` could not be imported.
  
- Fixed unit tests for rmagic when :mod:`numpy` is not installed, and
  for :mod:`numpy` is installed by :mod:`pandas` in missing.


Release 2.7.3
=============

Bugs fixed
----------

- method :meth:`DataFrame.collect` in :mod:`rpy2.robjects.lib.dplyr`
  was not functioning.

- Applied patch by Matthias Klose to fix implict pointer conversions.

- :mod:`pandas2ri.ri2py_dataframe` is now propagating the row names
  in the R data frame into an index in the pandas data frame (issue #285)

- methods `union`, `intersect`, `setdiff`, `ungroup` defined in the R package
  `dplyr` were missing from the
  `DataFrame` definition in :mod:`rpy2.robjects.lib.dplyr`


Release 2.7.2
=============

Bugs fixed
----------

- methods `distinct`, `sample_n`, and `sample_frac` defined in the R package
  `dplyr` were missing from the `DataFrame` definition in
  :mod:`rpy2.robjects.lib.dplyr`

- The fix for the inheritance problem with
  :mod:`rpy2.robjects.lib.dplyr.DataFrame` introduced a regression whenever
  `group_by` is used.

- The methods to perform joins on dplyr `DataFrame` objects where not
  working properly.


Release 2.7.1
=============
 
Bugs fixed
----------

- The :meth:`__repr__` for :mod:`robjects`-level vectors
  was broken for vectors of length 1 (issue #306)

- The ipython notebook-based sections of the documentation
  were not building

- Classes inheriting from :mod:`dplyr.DataFrame` had dplyr methods
  returning objects of their parent class.

Release 2.7.0
=============

New features
------------

- New exception :class:`rpy2.rinterface.RParsingError`. Errors
  occurring when parsing R code through :func:`rpy2.rinterface.parse`
  raise this exception (previously :class:`rpy2.rinterface.RRuntimeError`).

- New class :class:`rpy2.robjects.conversion.Converter` to replace
  the `namedtuple` of the same name

- New class :class:`rpy2.robjects.converter.ConversionContext`. This is
  a context manager allowing an easy setting of local conversion rules.
  The constructor has an alias called
  :meth:`rpy2.robjects.constructor.localconverter`.

- New module :mod:`rpy2.robjects.lib.dplyr` providing a custom
  wrapper for the R library `dplyr`

- Method :meth:`Environment.items()` to iterate through the symbols
  and associated objects in an R environment.

- Exception :class:`rpy2.rinterface.ParsingIncompleError`, a child class of
  :class:`rpy2.rinterface.ParsingError`, raised when
  calling :meth:`rpy2.rinteface.parse` results in R's C-level status
  to be `PARSE_INCOMPLETE`. This can make the Python implementation of an
  IDE for R easier.
  
- Attribute :attr:`slots` for :mod:`rpy2.robjects`-level objects. The
  attribute is a :class:`rpy2.robjects.Rslots` which
  behaves like a Python mapping to provide access to R-attributes
  for the object (see issue #275).

- The R "magic" for ipython `%%R` can be passed a local converter
  (see new features above) by using `-c`.


Bugs fixed
----------

- Conversion rules were not applied when parsing and evaluating string
  as R with :class:`rpy2.robjects.R`.

- Calling the constructor for :class:`rpy2.robjects.vectors.FactorVector`
  with an R factor is no longer making a copy, loosing the associated
  R attributes if any (fixes issue #299).

- `rpy2` could crash when R was unable to dynamically load the C extension for
  one of its packages (noticed with issue #303).
  

Changes
-------

- :func:`rpy2.rinterface.is_initialized` is now a function.

- :meth:`rpy2.robjects.R.__call__` is now calling R's `base::parse()`
  to parse the string rather the parser through R's C-API. The workaround
  let's us retrieve R's error message in case of failure (see issue #300)


Release 2.6.3
=============

Bug fixed
---------

- Metaclass `RS4Auto_Type` facilitating the creation of Python
  classes from R S4 classes was not handling classes without
  methods (issue #301)


Release 2.6.2
=============

Bugs fixed
----------
 
- Check that R >= 3.2 is used at build time (issue #291)

- Conversion rules were not applied when parsing and evaluating string
  as R code with :class:`rpy2.robjects.R`.


Release 2.6.1
=============

New features
------------

- Because of their long names, the classes
  :class:`SignatureTranslatedAnonymousPackage`,
  :class:`SignatureTranslatedPackage`, and
  :class:`SignatureTranslatedFunction`
  in :mod:`rpy2.robjects.packages` have now the aliases
  :class:`STAP`, :class:`STP`, and :class:`STF` respectively.

Bugs fixed
----------

- Typo in function name emitting warnings at build time (issue #283)
 
- The conversion of `TaggedList` instances is now handling the names
  of items in the list (issue #286)


Changes
-------

- Loading the `ipython` extension in the absence of `pandas` or `numpy`
  is now issuing a warning (issue #279)
  
  
Release 2.6.0
=============

New features
------------

- Report the existence during build time of a file `.Renviron`,
  or the definition of the environment variables `R_ENVIRON' or
  `R_ENVIRON_USER` with a warning. (issue #204)

- Moved console writting callback to use `ptr_R_WriteConsoleEx`
  rather than `ptr_R_WriteConsole`. This allows callbacks
  for warnings and messages. `get/set_writeconsole` is now 
  replaced by `get/set_writeconsole_regular` (regular
  output) and `get/set_writeconsole_warnerror` (warning and error).
  In order to conserve backward compatibility an alias for
  `get/set_writeconsole_regular` called `get/set_writeconsole` is
  provided.

- Added callback for `ptr_R_ResetConsole`.

- :mod:`pandas` :class:`Categorical` objects are automatically handled
  in the pandas converter.

- The translation of R symbols into Python symbols used in `importr` and
  underlying classes and methods can be customized with a callback.
  The default translation turning `.` into `_` is `default_symbol_r2python`.

- Translation of named arguments in R function is now sharing code with the
  translation of R symbols (see point above), providing a consistent way to
  perform translations.

- Utility function `sequence_to_vector` in `robjects` to convert Python
  sequences (e.g., `list` or `tuple`) to R vector without having to
  specify the type (the type is inferred from the list).

- :mod:`robjects.vectors` object have a property :attr:`NAvalue` that contains
  the `NA` value for the vector, allowing generic code on R vectors.
  For example, testing whether any vector contains `NA` can be written as
  `any(x is myvector.NAvalue for x in myvector)`. Making numpy /masked/ array
  is an other application.

Changes
-------

- The automatic name translation from R to Python used in `importr` is
  now slightly more complex. It will not only translate `.` to `_` but
  should a conflict arise from the existence in R of both the `.` and `_`
  versions the `.` version will be appended a `_` (in accordance with
  :pep:0008). The change was discussed in issue #274).

- The ipython 'R magic' is now starting with a default conversion mode
  that is `pandas2ri` if it can find it, then `numpy2ri` if it can find it,
  and then the basic conversion.

- R vectors are now typed at the C level (IntSexpVector, FloatSexpVector,
  ListSexpVector, etc...) whenever retrieving them from the embedded R
  with the low-level `rinterface`. This is facilitating dispatch on vector
  type (e.g., with `singledispatch` now used for the conversion system).
  
Bugs fixed
----------

- The evaluation of R code through R's C-level function `tryEval`
  caused console output whenever an error occurred. Moving to
  the seemingly experimental `tryEvalSilent` makes evaluations less
  verbose.

- Multiple plots in one ipython cell (pull request #44)

Release 2.5.7
=============

- `simplegeneric` was moved of ipython 4.0.0 (pull request #43)


Release 2.5.6
=============

Bugs fixed
----------

- Detection of the R version during setup on Win8 (issues #255 and #258)

- Segmentation fault when converting :mod:`pandas` :class:`Series` with
  elements of type object (issue #264)

- The default converter from Python (non-rpy2) objects to rinterface-level
  objects was producing robjects-level objects whenever the input was of
  type :class:`list` (discovered while fixing issue #264)

- Implemented suggested fix for issue with unlinking files on Windows
  (issue #191)

- Testing rpy2 in the absence of ipython no longer stops with an error
  (issue #266)


Release 2.5.5
=============

Bugs fixed
----------

- Crash (segfault) when querying an R object in an R environment triggers an
  error (symbol exists, but associated values resolves to an error - issue #251)

- Change in the signature of `rcall` was not updated in the documentation
  (issue #259)

- Minor update to the documentation (issue #257)


Release 2.5.4
=============

Bugs fixed
----------

- Filter PNG files on size, preventing empty files causing trouble to be
  ipython notebook rendering of graphics later on (slight modification of
  the pull request #39)

- Fix installation left unresolved with rpy2-2.5.3 (issue #248)

- Possible segfault with Python 3.4 (issue #249)


Release 2.5.3
=============

Changes
-------

- `setup.py` has `install_requires` in addition to `requires` in the hope to
   fix the missing dependency with Python 2 (:mod:`singledispatch` is required
   but not installed).

Bugs fixed
----------

- Extracting configuration information from should now work when R is emitting a warning (issue #247) 

- On OS X the library discovery step can yield nothing (see issue #246). A tentative fix is to issue
  a warning and keep moving.


Release 2.5.2
=============

Bugs fixed
----------

- String representation of :class:`robjects.R` (issue #238)

- Check during `build_ext` if unsupported version of R (pull request #32)

- HTMl display of columns of factors in a DataFrame (issue #236)

- HTML display of factors (issue #242)


Release 2.5.1
=============

Bugs fixed
----------

- Require singledispatch if Python 3.3 (issue #232)

- Fixed bug when R spits out a warning when asked configuration information (issue #233)

- Restored printing of compilation information when running `setup.py`

- Fixed installation issue on some systems (issue #234)

- Workaround obscure failure message from unittest if Python < 3.4 and
  :mod:`singledispatch` cannot be imported (issue #235)


Release 2.5.0
=============

New features
------------

- Experimental alternative way to preserve R objects from garbage collection.
  This can be activated with `rinterface.initr(r_preservehash=True)` (default
  is `False`.

- :class:`GGPlot` object getting a method :meth:`save`
  mirroring R's `ggplot2::ggsave()`.

- The conversion system is now using generics/single dispatch.

- New module :mod:`rpy2.ipython.html` with HTML display for rpy2 objects

- [Experimental] New function :func:`robjects.methods.rs4instance_factory`
  to type RS4 objects with more specificity.

Changes
-------

- The script `setup.py` was rewritten for clarity and ease of maintenance.
  Now it only uses `setuptools`.


Release 2.4.4
=============

Bugs fixed
----------

- Use `input` rather than `raw_input` in the default console callback
  with Python 3 (fixes issue #222)

- Issues with conversions, pandas, and rmagic (fixes issue #218 and more)


Release 2.4.3
=============

Bugs fixed
----------

- `geom_raster` was missing from `rpy2.robjects.lib.ggplot2` (pull request #30)

- Fixed issue with SVG rendering in ipython notebook (issue #217)

- Regression with `rx2()` introduced with new conversion (issue #219)

- Fixed documentation (missing `import`) (issue #213)


Release 2.4.2
=============

Bugs fixed
----------

- Assigning an R `DataFrame` into an environment was failing if
  the conversion for Pandas was activated. (Issue #207)


Release 2.4.1
=============

Bugs fixed
----------

- :meth:`rpy2.ipython` fixed spurious output to notebook cells.



Release 2.4.0
=============

Changes
-------

- Conversion system slightly changed, with the optional
  conversions for :mod:`numpy` and :mod:`pandas` modified
  accordingly. The changes should only matter if using
  third-party conversion functions.
 
- The Python 3 version is now a first class citizen. `2to3`
  is no longer used, and the code base is made directly
  compatible with Python. This lowers significantly the
  installation time with Python 3 
  (which matters when developping rpy2).

- The default options to initialize R (`rpy2.rinterface.initoptions') are no longer
  `('rpy2', '--quiet', '--vanilla', '--no-save')` but now
  `('rpy2', '--quiet', '--no-save')`.

- :class:`robjects.vectors.ListVector` can be instanciated from
  any objects with a method `items()` with the expectation that the method
  returns an iterable of (name, value) tuples, or even be an iterable
  of (name, value) tuples.

New features
------------

- For instances of :class:`rpy2.robjects.Function`,
  the `__doc__` is now a property fetching information
  about the parameters in the R signature.

- Convenience function :func:`rpy2.robjects.packages.data`
  to extract the datasets in an R pacakges

- :mod:`ipython`'s `rmagic` is now part of :mod:`rpy`. To use, `%load_ext
  rpy2.ipython` from within IPython.

- new method :meth:`rpy2.rinterface.SexpEnvironment.keys`, returnings
  the names in the environment as a tuple of Python strings.

- convenience class :class:`robjects.packages.InstalledPackages`, with a companion function
  :func:`robjects.packages.isinstalled`.

- new class :class:`rinterface.SexpSymbol` to represent R symbols

Bugs fixed
----------

- :meth:`rpy2.rinterface.Sexp.do_slot` was crashing when
  the parameter was an empty string (PR #155)



Release 2.3.10
==============

Bugs fixed
----------

- `setup.py build` was broken when new R compiled with OpenMP (Issue #183)

Release 2.3.9
=============

- Changes in pandas 0.13.0 broke the rpy2 conversion layer (Issue #173)


Release 2.3.8
=============

Bugs fixed
----------

- Crash with R-3.0.2. Changes in R-3.0.2's C API coupled to a strange behaviour
  with R promises caused the problem. (PR #150)


Release 2.3.7
=============

Bugs fixed
----------

- ggplot2's "guides" were missing

- ggplot2's "theme_classic" was missing (PR #143)

- ggplot2's "element_rect" was missing (PR #144)

- :func:`rpy2.interactive.packages` was broken (PR #142)


Release 2.3.6
=============

Bugs fixed
----------

- Several reports of segfault on OS X (since rpy2-2.3.1 - PR #109)

- More fixes in converting `DataFrames` with dates from `pandas`


Relase 2.3.5
============

Bugs fixed
----------

- Missing mapping to ggplot2's `scale_shape_discrete` function

- Better handling of dates in Pandas

- Constructor for POSIXct improved (and fixed)

Changes
-------

- The attribute :attr:`rclass` is no longer read-only and can be set
  (since R allows it)

- Importing the module :mod:`rpy2.interactive` no longer activates
  event processing by default (triggering concurrency errors
  when used with ipython).

New features
------------

- New module :mod:`rpy2.interactive.ipython` (so far plotting
  automatically a ggplot2 figure in the iPython's console)

- It is now possible to set the :attr:`rclass`.


Relase 2.3.4
============

Bugs fixed
----------

- Spurious error when running unit tests with Python 3 and numpy
  installed

- Missing mapping to ggplot2's `geom_dotplot` function

- Warnings are not longer printed (see Changes below)

Changes
-------

- Bumped target version of ggplot2 to 0.9.3.1

- Warnings are not longer printed. The C-level function in R became
  hidden in R-3.0, and the cost of an R-level check/print is relatively
  high if the R code called is very short. This might evolve into
  printing warnings only if interactive mode in Python (if this can
  be checked reliably).


Release 2.3.3
=============

Bugs fixed
----------

- Some of the data.frames converted from :mod:`pandas` were triggering
  a :class:`TypeError` when calling :func:`repr`

- In :mod:`rpy2.robjects.lib.ggplot2`, a mapping to `coord_fixed` was
  missing (PR #120)

- Using the parameter `lib_loc` in a call to
  :func:`rpy2.robjects.packages.importr` was resulting in an error (PR #119)

- Creating a `layer` through the `rpy2.robjects.lib.ggplot2` interface did
  not accept parameters (PR #122)

- Testing the Python version was crashing of a number of unsupported Python
  versions (<= 2.6) (PR #117)

New features
------------

- New module pandas2ri to convert from mod:`pandas` `DataFrame` objects

- New classes :class:`rpy2.robjects.lib.grid.Unit` and
  :class:`rpy2.robjects.lib.grid.Gpar` to model their counterparts in
  R's `grid` package as they were previously missing from rpy2.


Release 2.3.2
=============

Bug fixed
---------

- Building on Win64 (pull request #6)

- Fetching data from an R package through `importr` was masking
  any R object called `data` in that package. The data are now
  under the attribute name `__rdata__`. This is not completely
  safe either, although much less likely, a warning will
  be issued if still masking anything.
  

Changes
-------

- More informative error message when failing to build because `R CMD config`
  does not return what is expected

Release 2.3.1
=============

Bugs fixed
----------

- default console print callback with Python (issue #112 linked to it)

- deprecation warnings with ggplot2 (issue #111 and contributed patch)



Release 2.3.0
=============

New Features
------------

:mod:`rpy2.rinterface`:

- C-level API, allowing other C-level modules to make use of utilities
  without going through the Python level. The exact definition of
  the API is not yet fixed. For now there is 
  PyRinteractive_IsInitialized() to assess whether R was initialized
  (through :mod:`rpy2.rinterface` or not).

- C-module _rpy_device, allowing one to implement R graphical devices
  in Python [(very) experimental]

- Tracking of R objects kept protected from garbage collection by rpy2
  is now possible.

- New method :meth:`Sexp.rid` to return the identifier of the R object
  represented by a Python/rpy2 object

:mod:`rpy2.rinteractive`:

- Dynamic build of Python docstrings out of the R manual pages

:mod:`rpy2.robjects.help`:

- Build dynamic help

:mod:`rpy2.robjects.packages`:

- Build anonymous R packages from strings

- When using :func:`importr`, the datasets are added as an attribute
  :attr:`data`, itself an instance of a new class :class:`PackageData`.
  It no longer possible to access datasets are regular objects from
  a code package (because of changes in R), and the new system is
  more robust against quirks.

Changes
-------

:mod:`rpy2.rinterface`:

- :attr:`SexpClosure.env` to replace the method `closureenv`.


Release 2.2.6
=============

Bugs fixed
----------
 
- Newest R-2.15 and ggplot2 0.9 broke the ggplot2 interaface 
  in :mod:`rpy2.robjects.lib.ggplot2`


Release 2.2.5
=============

Bugs fixed
----------

- install process: Library location for some of the R installations

- should compile on win32 (thanks to a patch from Evgeny Cherkashin),
   a work to a limited extend


Release 2.2.4
=============

Bugs fixed
----------

- Memory leak when creating R vectors from Python (issue #82)

Release 2.2.3
=============

Bugs fixed
----------

- Dynamic construction of S4 classes was looking for R help as 'class.<class>'
  rather than '<class>-class'

- The cleanup of temporary directories created by R was not happening if
  the Python process terminated without calline :func:`rpy2.rinterface.endr()`
  (issue #68, and proof-of-principle fix by chrish42)


Release 2.2.2
=============

Bugs fixed
----------

- With the robjects layer, repr() on a list containing non-vector elements
  was failing

Release 2.2.1
=============

Bugs fixed
----------

- MANIFEST.in was missing from MANIFEST.in, required with Python 3


Release 2.2.0
=============

New Features
------------

- Support for Python 3, and for some of its features ported to Python 2.7

:mod:`rpy2.robjects`:

- :meth:`Environment.keys` to list the keys

- classes :class:`robjects.vectors.POSIXlt` and 
  :class:`robjects.vectors.POSIXlt` to represent vectors of R
  dates/time

- :func:`packages.get_packagepath` to get the path to an R package

- module :mod:`rpy2.robjects.help` to expose the R help system to Python

- Metaclass utilities in :mod:`rpy2.robjects.methods`, allowing to reflect
  automatically R S4 classes as Python classes.

- :meth:`rpy2.robjects.vectors.FactorVector.iter_labels` to iterate over the labels

- :class:`rpy2.robjects.vectors.ListVector` to represent R lists.

- Constructor for :class:`rpy2.robjects.vectors.ListVector` and 
  :class:`rpy2.robjects.vectors.DataFrame` accept any iterable at the condition
  that the elements iterated through also valid subscripts for it (e.g., given
  an iterable v, the following is valid:

  .. code-block:: python

     x[k] for x in v


:mod:`rpy2.rinterface`:

- :data:`NA_Complex` and :class:`NAComplexType` for missing complex values.

- :class:`SexpExtPtr` to represent R objects of type EXTPTR (external pointers).

- :func:`rpy2.rinterface.parse` to parse a string a R code

- :func:`rpy2.rinterface.rternalise` to wrap Python function as :class:`SexpClosure` that can
  be called by R just as it was a function of its own.

- :class:`rpy2.rinterface.RNULLType` for R's C-level NULL value and 
  :class:`rpy2.rinterface.UnboundValueType` for R's C-level R_UnboundValue
  (both singletons).

- :meth:`rinterface.SexpVector.index`, of similar behaviour to :meth:`list.index`.

- :meth:`rpy2.rinterface.Sexp.list_attrs` to list the names of all R attributes
  for a given object.

- :class:`rpy2.rinterface.ByteSexpVector` to represent R 'raw' vectors.

- constant `R_LEN_T_MAX` to store what is the maximum length for a vector in R.

- tuple `R_VERSION_BUILD` to store the version of R rpy2 was built against

- getter :attr:`Sexp.rclass` to return the R class associated with an object 

:mod:`rpy2.rlike`:

- :class:`container.OrdDict` get proper methods :meth:`keys` and `get`

:mod:`rpy2.interactive`:

- A new sub-package to provide utilities for interactive work, either for
  handling R interactive events or use Python for interactive programming
  (as often done with the R console)

Changes
-------

:mod:`rpy2.robjects`:

- NA_bool, NA_real, NA_integer, NA_character and NA_complex are now
  deprecated (and removed).
  NA_Logical, NA_Real, NA_Integer, NA_Character, NA_Complex should be used.

- :class:`rpy2.robjects.packages.Package` now inherits from :class:`types.ModuleType`

- classes representing R vector also inherit their type-specific
  rinterface-level counterpart.

- Importing the :class:`rpy2.robjects.numpy2ri` is no longer sufficient
  to active the conversion. Explicit activation is now needed; the function
  `activate` can do that.


:mod:`rpy2.rinterface`:

- :class:`IntSexpVector`, :class:`FloatSexpVector`,
  :class:`StrSexpVector`, :class:`BoolSexpVector`, :class:`ComplexSexpVector`
  are now defined at the C level, improving performances
  and memory footprint whenever a lot of instances are created.

Bugs fixed
----------

- Better and more explicit detection system for needed libraries when
  compiling rpy2 (ported to release 2.1.6)

- Long-standing issue with readline fixed (issue #10)


Release 2.1.9
=============

Bugs fixed
----------

- The R class in rpy2.robjects is now truly a singleton

- When using numpy 1.5 and Python >= 2.7, the exposed buffer for R numerical (double)
  vectors or arrays was wrong.


Release 2.1.8
=============

Bugs fixed
----------

- Fixed issue with R arrays with more than 2 dimensions and numpy arrays
  (issue #47 - backported from the branch 2.2.x).


Release 2.1.7
=============

Bugs fixed
----------

- More fixes for the automated detection of include and libraries at build time.


Release 2.1.6
=============

Bugs fixed
----------

- Further fixes in the automatic detection of includes and libraries
  needed to compile rpy2 against R. The detection code has
  been refactored (backport from the 2.2.x branch)


Release 2.1.5
=============

Bugs fixed
----------

- fixes the automatic detection of R_HOME/lib during building/compiling
  when R_HOME/lib is not in lib/ (issue #54)


Release 2.1.4
=============

New features
------------

- :mod:`rpy2.robjects.lib.ggplot2` now has the functions :func:`limits`,
  :func:`xlim`, :func:`ylim` exposed (patch contributed anonymously)

Bugs fixed
----------

- Install script when the BLAS library used by R is specified as a library
  file (patch by Michael Kuhn)


Release 2.1.3
=============

Bugs fixed
----------

- Spurious error message when using DataFrame.from_csvfile() without
  specifying col_names  or row_names

- Patch to finally compile with Python < 2.6 (contribDuted by Denis Barbier)


Release 2.1.2
=============

New Features
------------

:mod:`rpy2.robjects`:

- NA_Logical, NA_Real, NA_Integer, NA_Character from :mod:`rpy2.rinterface` 
  are imported by robjects.

Changes
-------

:mod:`rpy2.robjects`:

- NA_bool, NA_real, NA_integer, NA_character and NA_complex are now
  robjects-level vectors
  (they were rinterface-level vectors).
  Consider using the rinterface-defined NAs instead of them.

Bugs fixed
----------

- Missing conditional C definition to compile with Python 2.4 # issue 38

- Fixed error when calling robjects.vectors.Vector.iteritems() on an R 
  vector without names

- Fixed automatic conversion issues (issue #41)


Release 2.1.1
=============

Bugs fixed
----------

- Issues with NA values # issue 37

- Missing manual scale functions in :mod:`rpy2.robjects.lib.ggplot2` # issue 39


Release 2.1.0
=============

New Features
------------

:mod:`rpy2.robjects`:

- Method :meth:`formals` for :class:`Function` (formerly *RFunction*)

- Methods :meth:`slotnames`, :meth:`isclass`, and :meth:`validobject`
  for :class:`RS4`

- Vector-like objects now in a module :mod:`rpy2.robjects.vectors`

- :func:`set_accessors` for adding simply accessors to a class inheriting
  from :class:`RS4`

- :class:`RS4_Type` for metaclass-declared accessors

- Delegating classes :class:`ExtractDelegator` and 
  :class:`DoubleExtractDelegator` for extracting the R-way

- :class:`DataFrame` (formerly *RDataFrame*) can now be created 
  from :`rlike.container.OrdDict`
  instances, or any other object inheriting from dict.

- :class:`FactorVector` to represent R factors

- the conversion is now returning subclasses of
  :class:`robjects.vectors.Vector` -formerly *RVector*-
  (such as :class:`IntVector`, 
  :class:`FloatVector`, etc...) rather than only return :class:`Vector`

- :class:`StrVector` has a method :meth:`factor` to turn a
  vector of strings into an R factor

- :class:`Matrix` was added the methods: :meth:`dot`, :meth:`svd`, :meth:`crossprod`,
  :meth:`tcrossprod`, :meth:`transpose`.
 
- :meth:`IntVector.tabulate` to count the number of times a value is found in the vector

- :meth:`Vector.sample` to draw a (random) sample of arbitrary size from a vector

- :data:`NA_Bool`, :data:`NA_Real`, :data:`NA_Integer`, :data:`NA_Character`, 
  :data:`NA_Complex` as aliases for R's missing values.

- :data:`ComplexVector` for vectors of complex (real + imaginary) elements

- :mod:`packages` to provide utility functions to handle R packages
  (import of R packages)

- :mod:`functions` to provide classes related to R functions, with the new 
  class :class:`SignatureTranslatedFunction`

- :meth:`DataFrame.iter_row` and :meth:`DataFrame.iter_column`, iterating
  through rows and columns respectively.

- :meth:`DataFrame.cbind` and :meth:`DataFrame.rbind` for binding columns or
  rows to a DataFrame.

- :meth:`Vector.iteritems` to iterate on pairs of names and values.

- :attr:`Robject.__rname__` to store the "R name"

:mod:`rpy2.rinterface`:

- New functions for specifying callback functions for R's front-ends:
  :func:`set_showmessage`, :func:`set_flushconsole`,
  :func:`set_choosefile`, :func:`set_showfiles`

- New object :data:`MissingArg`, exposing R's special object for representing
  a "missing" parameter in a function call.
  (#this was first a patch by Nathaniel Smith with a function getMissingArgSexp)

- Initial commit of a callback-based implementation of an R graphical device
  (this is for the moment very experimental - and not fully working)

- :meth:`SexpClosure.rcall` is now taking 2 parameters, 
  a tuple with the parameters and
  an :class:`SexpEnvironment` in which the call is to be evaluated.

- :attr:`Sexp.__sexp__` now has a setter method. This permits the rebinding
  of the underlying R SEXP, and allows to expose `foo<-` type of R methods
  as Python function/methods with side effects.

- Objects of R type RAWSXP are now exposed as instances of class
  :class:`SexpVector`.

- Factory function :func:`unserialize` to build Sexp* instances from byte
  string serialized with R's own 'serialize'.

- Method :meth:`Sexp.__reduce__` for pickling/unpickling

- Ability to specify a callback function for R_CleanUp (called upon exiting R)
  through :func:`get_cleanup` and :func:`set_cleanup` [very experimental]

- Class :class:`ListSexpVector` for creating R lists easily
  (complementing :class:`IntSexpVector`, :class:`StrSexpVector`, and friends) 

- :meth:`colnames`, :meth:`rownames` for :class:`Array` (formerly *RArray*) are now 
  property-style getters

- Pairlists (LISTSXP) now handled

- Experimental function :func:`set_interactive` to set whether R is in interactive
  mode or not (#following an issue reported by Yaroslav Halchenko)

- New object :data:`R_NilValue`, exposing R's special object for representing
  a "NULL".

- :data:`ComplexSexpVector` for vectors of complex (real + imaginary) elements

- Scalar Python parameters of type :class:`int`, :class:`long`,
  :class:`double`, :class:`bool`, and :class:`None` 
  in a call (using :class:`SexpClosure`) are now automatically converted
  to length-one R vectors (at the exception of None, converted
  to R_NilValue).

- Python slices can now be used on R vector-like objects 

- Better handling of R's missing values NA, `NA_integer_`, `NA_real_`, 
  and `NA_character_`.

:mod:`rpy2.rlike`:

- :meth:`iteritems` for :class:`OrdDict` (formerly:class:`ArgDict`) 
  and :class:`TaggedList`

- static method :meth:`from_iteritems` for :class:`TaggedList`,
  for creating a TaggedList from any object having a method :meth:`iteritems`


Changes
-------

- The setup.py script is now taking command-line arguments when specifying
  R library-related paths are wished. python setup.py --help build_ext will
  list them
  
:mod:`rpy2.robjects`:

- RS4 no longer makes R's slots as Python attributes through :meth:`__attr__`

- The package is split into modules

- The broken variables NA_STRING, NA_INTEGER, NA_LOGICAL, and NA_REAL are
  removed. The documentation on missing values was revised.

- :data:`globalEnv` and :data:`baseNameSpaceEnv` were renamed to 
  :data:`globalenv` and :data:`baseenv` respectively

- The parameter *wantFun* in :meth:`Environment.get` 
  (formerly *REnvironment.get()*) is now *wantfun* 

- :attr:`Vector.r` does not have a __getitem__ method any longer
  (see in `.rx` and `.rx2` in the new features)

- :meth:`colnames`, :meth:`rownames`, :meth:`nrow`, :meth:`ncol` 
  for :class:`DataFrame` are now property-style getters

- :meth:`nrow`, :meth:`ncol` for :class:`Array` 
  are now property-style getters

- static method :meth:`from_csvfile` and 
  instance method :meth:`to_csvfile` for :class:`DataFrame`

- module :mod:`lib` to store modules representing R packages

- module :mod:`lib.ggplot2` for the CRAN package ggplot2.

- renaming of few classes, the *R* prefix: :class:`Formula` (from *RFormula*),
  :class:`DataFrame` (from *RDataFrame*), :class:`Array` (from *RArray*),
  :class:`Matrix` (from *RMatrix*), :class:`Environment` (from *REnvironment*),
  :class:`Function` (from *RFunction*), :class:`Vector` (from *RVector*).

- :class:`robjects.vectors.Vector` lost the (now redundant) methods
  `subset` and `assign`. Those operations were just aliases to the 
  :class:`ExtractDelegator`

:mod:`rpy2.rinterface`:

- :data:`globalEnv`, :data:`baseNameSpaceEnv`, and :data:`emptyEnv` 
  were renamed to :data:`globalenv`, :data:`baseenv` and :data:`emptyenv`
  respectively

- The parameter *wantFun* in :meth:`SexpEnvironment.get` is now *wantfun* 

- The call-back getters and setters are now :func:`get_readconsole`,
  :func:`set_readconsole`, :func:`get_writeconsole`, :func:`set_writeconsole`,
  :func:`get_flushconsole`, and :func:`set_flushconsole`.

- Functions also accept named parameters equal to Py_None, and transform them to R
  NULL (previously only accepted parameters inheriting from Sexp).

:mod:`rpy2.rlike`:

- :class:`ArgDict` becomes :class:`OrdDict`.

- :meth:`tags` of :class:`TaggedList` is now a property (with a getter
  and a setter)

:mod:`rpy2.rpy_classic`:

- R named lists are returned as Python :class:`dict`, like rpy-1.x does it, with the
  notable difference that duplicate names are not silently overwritten: an exception
  of class :class:`ValueError` is thrown whenever happening


Bugs fixed
----------

- :meth:`REnvironment.get` now accepts a named parameter *wantFun*
  (like :meth:`rinterface.SexpEnvironment` does)

- :class:`rinterface.SexpVector` will now properly raise an exception
  when trying to create vector-like object of impossible type

- Crash when trying to create a SexpVector of a non-vector type

- R objects of class *matrix* are now properly converted into :class:`RMatrix`
  (instead of :class:`Array`)

- :meth:`Robj.as_py` was not working at all (and now it does to some extent)


Release 2.0.7
=============

Bugs fixed
----------

- On win32, printing an object was leaving an open file handle behind each time,
  leading to an error and the impossibility to print
  (# bug report and fix by Christopher Gutierrez)


Release 2.0.6
=============

No user-visible change. Win32-specific additions to the C module
were made to compile it.


Release 2.0.5
=============

Bugs fixed
----------

- Crash when calling :meth:`SexpEnvironment.get` with an empty string
  #bug report by Walter Moreira

- :meth:`SexpEnvironment.__getitem__` called with an empty string
  caused unpredictable (and bad) things


Release 2.0.4
=============

Bugs fixed
----------

- Added missing named parameter *wantfun* to method :meth:`REnvironment.get`
  (making it similar to :meth:`SexpEnvironment.get`)

- Leak in reference counting when creating SexpVector objects fixed
  (the symptom was a process growing in size when creating R vector from
  Python list or numpy arrays)

- `R CMD config LAPACK_LIBS` could return an empty string when R was compiled
  with the veclib framework, causing the setup.py script to raise an exception.
  setup.py now only print a message about an empty string returned from 
  R CMD config

- Numpy arrays with complex elements are no longer causing segfaults

- Calls to :meth:`SexpClosure.rcall` with something else that the expected
  kind of tuple could cause a segfault


Release 2.0.3
=============

New Features
------------

:mod:`rpy2.rinterface`:

- :meth:`process_revents`, a Wrapper for R_ProcessEvents 
  (# suggested by June Kim to help with issues related to interactive display on win32),
  and for R_RunHandlers on UNIX-like systems 
  (# patch by Nathaniel Smith).

- All callbacks are getting a get<callback> to complement the set<callback>.
  (# Patch by Nathaniel Smith) 

- :meth:`Sexp.__deepcopy__` to copy an object (calling Rf_Duplicate) 
   (# from a patch by Nathaniel Smith)
 
Changes
-------
 
- the default for reading and writing the console are now using sys.stdin and sys.stdout 
  (# patch submitted by Nathaniel Smith)
 
- console IO callbacks (reading and writing) are complemented by 
  one to flush the console

- :meth:`Sexp.do_slot_assign` now creates the slot if missing
  (design-fix - # patch by Nathaniel Smith)
 

Bugs fixed
----------
 
- fixed problem of numpy interface with R boolean vectors. 
  They are now presented as 'i' rather than 'b' to numpy
  (# patch submitted by Nathaniel Smith)
 
- The mechanism for setting arbitrary callaback functions for console I/O 
  now ensures that a traceback is printed to stderr whenever an error
  occurs during the evalutation of the callback (the raised exception used
  to be silently propagated to the next python call, leading to problems).



Release 2.0.2
=============

Bugs fixed
----------

- Fix installation bug when the include directories contain either '-' or 'I'
  #spotted by James Yoo

- Failing to initialize R now throws a RuntimeError

- Copying an R "NA" into Python returns a None (and no longer a True)
  (#fixes a bug reported by Jeff Gentry)


Release 2.0.1
=============

New features
------------

:mod:`rpy2.robjects`:

- Property `names` for the :class:`RVector` methods :meth:`getnames` 
  and :meth:`setnames` (this was likely forgotten for Release 2.0.0).

- Property `rclass` for :class:`RObjectMixin` 

Changes
-------

:mod:`rpy2.robjects`:

- :meth:`rclass` becomes :meth:`getrclass`

Bugs fixed
----------

- Having the environment variable R_HOME specified resulted in an error
  when importing :mod:`rpy2.rinterface` # root of the problem spotted by Peter

- Setup.py has no longer a (possibly outdated) static hardcoded version number
  for rpy2

- Testing no longer stops with an error in the absence of the third-party
  module :mod:`numpy`

- :meth:`rpy2.rlike.container.TaggedList.pop` is now returning the element
  matching the given index



Release 2.0.0
=============

New features
------------

- New module :mod:`rpy2.robjects.conversion`.

- New module :mod:`rpy2.robjects.numpy2ri` to convert :mod:`numpy` objects
  into :mod:`rpy2` objects. 
  # adapted from a patch contributed by Nathaniel Smith


Changes
-------

- :meth:`RObject.__repr__` moved to :meth:`RObject.r_repr`


Bugs fixed
----------

- Informative message returned as RuntimeError when failing to find R's HOME

- Use the registry to find the R's HOME on win32
  # snatched from Peter's earlier contribution to rpy-1.x


Release 2.0.0rc1
================

:mod:`rpy2.rpy_classic`:

- :meth:`rpy_classic.RObj.getSexp` moved to a 
  property :attr:`rpy_classic.Robj.sexp`.

:mod:`rpy2.robjects`:

- :meth:`RObject.__repr__` moved to :meth:`RObject.r_repr`

- :meth:`ri2py`, :meth:`ro2py`, and :meth:`py2ri` moved to the new module
  :mod:`conversion`. Adding the prefix `conversion.` to calls
  to those functions will be enough to update existing code


Bugs fixed
----------

- Informative message returned as RuntimeError when failing to find R's HOME

- Use the registry to find the R's HOME on win32
  # snatched from Peter's earlier contribution to rpy-1.x


Release 2.0.0rc1
================

New features
------------

- added :data:`__version__` to rpy2/__init__.py

:mod:`rpy2.robjects`:

- added classes :class:`StrVector`, :class:`IntVector`, :class:`FloatVector`, :class:`BoolVector`

:mod:`rpy2.rinterface`:

- added missing class :class:`BoolSexpVector`.


Changes
-------

:mod:`rpy2.robjects`:

- does not alias :class:`rinterface.StrSexpVector`, :class:`rinterface.IntSexpVector`, :class:`rinterface.FloatSexpVector` anymore

- Constructor for :class:`rpy2.robjects.RDataFrame` checks that R lists are data.frames (not all lists are data.frame)

- Formerly new attribute :attr:`_dotter` for :class:`R` is now gone. The documentaion now points to :mod:`rpy2.rpy_classic` for this sort of things.

Bugs fixed
----------

- conditional typedef in rinterface.c to compile under win32 # reported and initial proposed fix from Paul Harrington

- __pow__ was missing from the delegator object for robjects.RVector (while the documentation was claiming it was there) # bug report by Robert Nuske

- Earlier change from Sexp.typeof() to getter Sexp.typeof was not reflected in :mod:`rpy2.rpy_classic` # bug report by Robert Denham

 
Release 2.0.0b1
===============

New features
------------

:mod:`rpy2.robjects`:

- added :meth:`setenvironment` for :class:`RFormula`, and defined `environment` as a property

- defined `names` as a property for :class:`RVector`

:mod:`rpy2.rinterface`:

- added functions :func:`get_initoptions` and :func:`set_initoptions`.

- new attribute :attr:`_dotter` for :class:`R` singleton. Setting it to True will translate '_' into '.' if the attribute is not found 

Changes
-------

:mod:`rpy2.robjects`:

- constructor for RDataFrame now now accepts either :class:`rlike.container.TaggedList` or :class:`rinterface.SexpVector`

:mod:`rpy2.rinterface`:

- :func:`sexpTypeEmbeddedR` is now called :func:`str_typeint`.

- :attr:`initOptions` is now called :attr:`initoptions`. Changes of options can only be done through :func:`set_initoptions`.


Bugs fixed
----------

- crash of :meth:`Sexp.enclos` when R not yet initialized (bug report #2078176)

- potential crash of :meth:`Sexp.frame` when R not yet initialized

- proper reference counting when handling, and deleting, :attr:`Sexp.__sexp__` generated CObjects

- setup.py: get properly the include directories (no matter where they are) #bug report and fix adapted from Robert Nuske

- setup.py: link to external lapack or blas library when relevant

- added a MANIFEST.in ensuring that headers get included in the source distribution #missing headers reported by Nicholas Lewin-Koh

- :func:`rinterface.str_typeint` was causing segfault when called with 99

- fixed subsetting for LANGSXP objects


Release 2.0.0a3
===============

New features
------------

:mod:`rpy2.rinterface`:

- :func:`setReadConsole`: specify Python callback for console input

- `R` string vectors can now be built from Python unicode objects

- getter :attr:`__sexp__` to return an opaque C pointer to the underlying R object

- method :meth:`rsame` to test if the underlying R objects for two :class:`Sexp` are the same.

- added `emptyEnv` (R's C-level `R_EmptyEnv`)

- added method :meth:`Sexp.do_slot_assign`

:mod:`rpy2.robjects`:

- R string vectors can now be built from Python unicode objects

:mod:`rpy2.rlike`:

- module :mod:`functional` with the functions :func:`tapply`, :func:`listify`, :func:`iterify`.

- module :mod:`indexing` with the function :func:`order`

- method :meth:`TaggedList.sort` now implemented

 
Changes
-------

:mod:`rpy2.rinterface`:

- :func:`initEmbeddedR` is only initializing if R is not started (no effect otherwise, and no exception thrown anymore)

- the method :meth:`Sexp.typeof` was replaced by a Python `getter` :attr:`typeof`.

- the method :meth:`Sexp.named` was replaced by a Python `getter` :attr:`named`.

- R objects of type LANGSXP are now one kind of vector (... but this may change again)

- R objects of type EXPRSXP are now handled as vectors (... but this may change again)

- :func:`initEmbeddedR` renamed to :func:`initr`

- :func:`endEmbeddedR` renamed to :func:`endr`


:mod:`rpy2.robjects`:

- :class:`R` remains a singleton, but does not throw an exception when multiple instances are requested


Bugs fixed
----------

- unable to compile on Python2.4 (definition of aliases to Python2.5-specific were not where they should be).

- overflow issues on Python 2.4/64 bits when indexing R vector with very large integers.

- handling of negative indexes for :class:`SexpVector`'s :meth:`__getitem__` and :meth:`__setitem__` was missing

- trying to create an instance of :class:`SexpVector` before initializing R raises a RuntimeException (used to segfault)

- experimental method :meth:`enclos` was not properly exported

- setup.py was exiting prematurely when R was compiled against an existing BLAS library

- complex vectors should now be handled properly by :mod:`rpy2.rinterface.robjects`.

- methods :meth:`rownames` and :meth:`colnames` for :class:`RDataFrame` were incorrect.

Release 2.0.0a2
===============

New features
------------

:mod:`rpy2.rlike`:

- package for R-like features in Python

- module :mod:`rpy2.rlike.container` 

- class :class:`ArgsDict` in :mod:`rpy2.rlike.container`

- class :class:`TaggedList` in :mod:`rpy2.rlike.container`

:mod:`rpy2.rinterface`:

- method :meth:`named`, corresponding to R's C-level NAMED

- experimental methods :meth:`frame` and :meth:`enclos` for SexpEnvironment corresponding to R's C-level FRAME and ENCLOS

- method :meth:`rcall` for :class:`ClosureSexp`

- new experimental class :class:`SexpLang` for R language objects.


Bugs fixed
----------

- R stack checking is disabled (no longer crashes when multithreading)

- fixed missing R_PreserveObject for vectors (causing R part of the object to sometimes vanish during garbage collection)

- prevents calling an R function when R has been ended (raise :class:`RuntimeException`).


Release 2.0.0a1
===============

New features
------------

:mod:`rpy2.robjects`:

- method :meth:`getnames` for :class:`RVector`

- experimental methods :meth:`__setitem__` and :meth:`setnames` for :class:`RVector`

- method 'getnames' for :class:`RArray`

- new class :class:`RFormula`

- new helper class :class:`RVectorDelegator` (see below)

- indexing RVector the "R way" with subset is now possible through a delegating attribute (e.g., myvec.r[True] rather than myvec.subset(True)). #suggested by Michael Sorich

- new class :class:`RDataFrame`. The constructor :meth:`__init__` is still experimental (need for an ordered dictionnary, that will be in before the beta

- filled documentation about mapping between objects

Changes
-------

- many fixes and additions to the documentation

- improved GTK console in the demos

- changed the major version number to 2 in order to avoid confusion with rpy 1.x # Suggested by Peter and Gregory Warnes

- moved test.py to demos/example01.py

:mod:`rpy2.robjects`:

- changed method name `getNames` to `getnames` where available (all lower-case names for methods seems to be the accepted norm in Python).


Bugs fixed
----------

:mod:`rpy2.robjects`:

- fixed string representation of R object on Microsoft Windows (using fifo, not available on win32)

- :meth:`__getattr__` for :class:`RS4` is now using :meth:`ri2py` 

:mod:`rpy2.rinterface`:

- fixed context of evaluation for R functions (now R_GlobalEnv)

Release 1.0a0
=============

- first public release