Release 2.1 (#231)

* Release 2.1

CHANGELOG.md

@@ -1,7 +1,22 @@
-## master
-    - Support for chunk cache configuration at data set create/open
-    - Support for std::array (unnested)
-    - Added AnnotateTraits::createAttribute and NodeTraits::createDataset overloads to create and initialize the data in a single call
+## Version 2.1 - 2019/10/XX
+### New Features:
+    - Inspection: API to get the type of links/objects and datasets data-types (#221)
+    - H5Easy: API for simple import/export to Eigen and xtensor (#141)
+    - Support for chunk and deflate configuration at dataset creation/open (#125). Added generic RawPropertyLists. (#157)
+    - Recursive `createGroup` and `exist` (#152)
+    - Shortcut syntax: ability to create a filled dataset in a single line (#130)
+    - DataSet now accepts `std::complex` and `std::array`'s (#128, #129)
+
+### Improvements
+    - Improved compat with MSVC and ICC compilers
+    - CMake build system: modernized, create exported targets, better messages, etc.
+    - Building and publishing documentation: https://bluebrain.github.io/HighFive/
+    - Several other. See #231
+
+### Bug Fixes:
+    - Fixed header dependencies. They are now all include-able (#225)
+    - Fixed read/write of N-Dimensional data as nested vectors (#191)
+    - Fixed data broadcasting for reading (#136)
 
 ## Version 2.0 - 2018/07/19
     - First version with C++11 enforcement
@@ -29,7 +44,6 @@
     - Introduce SilenceHDF5 for HDF5 error report
     - Fix a unit test issue with SilenceHDF5
 
-
 ## Version 1.3 - 2017/06/21
     - Minor fixes
 
@@ -46,6 +60,5 @@
     - Improve documentation
     - Add example for boost.Ublas matrix support
 
-
 ## Version 1.0 - Init
 	- Initial release

CMakeLists.txt

@@ -5,7 +5,7 @@ else()
   cmake_policy(VERSION 3.13)
 endif()
 
-project(HighFive VERSION 2.0)
+project(HighFive VERSION 2.1)
 
 list(APPEND CMAKE_MODULE_PATH
   ${PROJECT_SOURCE_DIR}/CMake

README.md

@@ -21,9 +21,12 @@ HighFive does not require an additional library and supports both HDF5 thread sa
 
 
 ### Dependencies
-
 - libhdf5
-- boost >= 1.41 (optional, disable in CMake using `-DUSE_BOOST=FALSE`)
+- (optional, opt-in with *HIGHFIVE_PARALLEL_HDF5*) libhdf5-mpi
+- (optional, opt-out with *USE_BOOST*) boost >= 1.41
+- (optional, opt-in with *USE_EIGEN*) eigen3
+- (optional, opt-in with *USE_XTENSOR*) xtensor
+
 
 ### CMake integration
 
@@ -36,14 +39,9 @@ cmake_minimum_required(VERSION 3.0 FATAL_ERROR)
 project(foo)
 set(CMAKE_CXX_STANDARD 11)
 
-find_package(HighFive 2.0 REQUIRED)
+find_package(HighFive 2.1 REQUIRED)
 add_executable(bar bar.cpp)
-target_include_directories(
-  bar
-  PUBLIC $<TARGET_PROPERTY:HighFive,INTERFACE_INCLUDE_DIRECTORIES>)
-target_link_libraries(
-  bar
-  PUBLIC $<TARGET_PROPERTY:HighFive,INTERFACE_LINK_LIBRARIES>)
+target_link_libraries(bar HighFive)
 ```
 
 ### Usage
@@ -143,6 +141,7 @@ whereby the `int` type of this example can be replaced by any of the above types
 
 ### Test Compilation
 Remember: Compilation is not required. Used only for unit test and examples
+Unit tests need boost (*USE_BOOST*).
 
 ```bash
 mkdir build; pushd build
@@ -153,7 +152,7 @@ make test
 
 ### Feature support
 
-- create/read/write file,  dataset, group, dataspace.
+- create/read/write file, dataset, group, dataspace.
 - automatic memory management / ref counting
 - automatic conversion of `std::vector` and nested `std::vector` from/to any dataset with basic types
 - automatic conversion of `std::string` to/from variable length string dataset
@@ -171,6 +170,7 @@ make test
 - Tristan Carel <tristan.carel@epfl.ch> - Blue Brain Project
 - Wolf Vollprecht <w.vollprecht@gmail.com> - QuantStack
 - Tom de Geus <tom@geus.me> - EPFL
+- Nicolas Cornu <nicolas.cornu@epfl.ch> - Blue Brain Project
 
 ### License
 

VERSION

@@ -1 +1 @@
-2.0
+2.1