Add Blaze integration

closes #2

README.Blaze.md

@@ -0,0 +1,41 @@
+# Blaze (blaze-lib) Matrix Market
+
+`fast_matrix_market` provides read/write methods for [Blaze](https://bitbucket.org/blaze-lib/blaze) sparse and dense matrices and vectors.
+
+# Usage
+
+```c++
+#include <fast_matrix_market/app/Blaze.hpp>
+```
+
+```c++
+blaze::CompressedMatrix<double> A;
+// or
+blaze::DynamicMatrix<double> A;
+// or
+blaze::CompressedVector<double> A;
+// or
+blaze::DynamicVector<double> A;
+```
+### Reading
+```c++
+std::ifstream f("input.mtx");
+
+fast_matrix_market::read_matrix_market_blaze(f, A);
+```
+
+**Note:** if `A` is a Vector type then the Matrix Market file must be either a vector or a
+1-by-N row or M-by-1 column matrix. Any other matrix will cause an exception to be thrown.
+
+### Writing
+
+```c++
+std::ofstream f("output.mtx");
+
+fast_matrix_market::write_matrix_market_blaze(f, A);
+```
+
+
+**Compatibility Note:** Blaze Vector objects are written as vector Matrix Market files.
+Many MatrixMarket loaders cannot read vector files. If this is a problem write the vector
+as a 1-by-N or M-by-1 matrix.

README.md

@@ -113,6 +113,13 @@ cs_dl *A;
 fast_matrix_market::read_matrix_market_cxsparse(input_stream, &A, cs_dl_spalloc);
 ```
 
+## Blaze
+[Blaze](https://bitbucket.org/blaze-lib/blaze) sparse and dense matrices and vectors are supported. See [Blaze README](README.Blaze.md).
+```c++
+blaze::CompressedMatrix<double> A;
+fast_matrix_market::read_matrix_market_blaze(input_stream, A);
+```
+
 ## Your Own
 
 Simply provide `parse_handler` and `formatter` classes to read and write from/to any datastructure, respectively. The class you need is likely already in the library.

cmake/Blaze.cmake

@@ -0,0 +1,10 @@
+# Fetch from git
+include(FetchContent)
+
+FetchContent_Declare(
+        blaze
+        GIT_REPOSITORY https://bitbucket.org/blaze-lib/blaze.git
+        GIT_TAG v3.8.1
+        GIT_SHALLOW TRUE)
+
+FetchContent_MakeAvailable(blaze)