Merge c1b7c10 into 775ee5d

LLNL · Jan 6, 2021 · f211a12 · f211a12
2 parents 775ee5d + c1b7c10
commit f211a12
Show file tree

Hide file tree

Showing 9 changed files with 629 additions and 2 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -17,6 +17,8 @@ and this project aspires to adhere to [Semantic Versioning](https://semver.org/s
 #### General 
 - Added a builtin sandboxed header-only version of fmt. The namespace and directory paths were changed to `conduit_fmt` to avoid potential symbol collisions with other codes using fmt. Downstream software can use by including `conduit_fmt/conduit_fmt.h`.
 - Added support for using C++11 initializer lists to set Node and DataArray values from numeric arrays. See C++ tutorial docs (https://llnl-conduit.readthedocs.io/en/latest/tutorial_cpp_numeric.html#c-11-initializer-lists) for more details.
+- Added a Node::describe() method. This method creates a new node that mirrors the current Node, however each leaf is replaced by summary stats and a truncated display of the values. For use cases with large leaves, printing the describe() output Node is much more helpful for debugging and understanding vs wall of text from other to_string() methods. 
+
 
 #### Relay
 - Added Relay IO Handle mode support for `a` (append) and `t` (truncate).  Truncate allows you to overwrite files when the handle is opened. The default is append, which preserves prior IO Handle behavior.

diff --git a/src/libs/conduit/conduit_data_array.cpp b/src/libs/conduit/conduit_data_array.cpp
@@ -279,6 +279,81 @@ DataArray<T>::diff_compatible(const DataArray<T> &array, Node &info, const float
     return res;
 }
 
+//---------------------------------------------------------------------------// 
+///
+/// Summary Stats Helpers
+///
+//---------------------------------------------------------------------------// 
+
+//---------------------------------------------------------------------------// 
+template <typename T>
+T
+DataArray<T>::min()  const
+{
+    T res = std::numeric_limits<T>::max();
+    for(index_t i = 0; i < number_of_elements(); i++)
+    {
+        const T &val = element(i);
+        if(val < res)
+        {
+            res = val;
+        }
+    }
+
+    return res;
+}
+
+//---------------------------------------------------------------------------// 
+template <typename T>
+T
+DataArray<T>::max()  const
+{
+    T res = std::numeric_limits<T>::min();
+    for(index_t i = 0; i < number_of_elements(); i++)
+    {
+        const T &val = element(i);
+        if(val > res)
+        {
+            res = val;
+        }
+    }
+
+    return res;
+}
+
+
+//---------------------------------------------------------------------------// 
+template <typename T>
+T
+DataArray<T>::sum()  const
+{
+    T res =0;
+    for(index_t i = 0; i < number_of_elements(); i++)
+    {
+        const T &val = element(i);
+        res += val;
+    }
+
+    return res;
+}
+
+//---------------------------------------------------------------------------// 
+template <typename T>
+float64
+DataArray<T>::mean()  const
+{
+    float64 res =0;
+    for(index_t i = 0; i < number_of_elements(); i++)
+    {
+        const T &val = element(i);
+        res += val;
+    }
+
+    res = res / float64(number_of_elements());
+    return res;
+}
+
+
 //---------------------------------------------------------------------------// 
 template <typename T>
 std::string 
@@ -1436,6 +1511,133 @@ DataArray<T>::compact_elements_to(uint8 *data) const
 }
 
 
+//---------------------------------------------------------------------------//
+template <typename T> 
+std::string
+DataArray<T>::to_summary_string_default() const
+{ 
+    return to_summary_string();
+}
+
+//---------------------------------------------------------------------------//
+template <typename T> 
+std::string
+DataArray<T>::to_summary_string(index_t threshold) const
+{ 
+    std::ostringstream oss;
+    to_summary_string_stream(oss, threshold);
+    return oss.str();
+}
+
+//---------------------------------------------------------------------------//
+template <typename T> 
+void
+DataArray<T>::to_summary_string_stream(std::ostream &os,
+                                       index_t threshold) const
+{ 
+    // if we are less than or equal to threshold, we use to_yaml
+    index_t nele = number_of_elements();
+
+    if(nele <= threshold)
+    {
+        to_yaml_stream(os);
+    }
+    else
+    {
+        // if above threshold only show threshold # of values
+        int half = threshold / 2;
+        int bottom = half;
+        int top = half;
+
+        //
+        // if odd, show 1/2 +1 first
+        //
+
+        if( (threshold % 2) > 0)
+        {
+            bottom++;
+        }
+
+        if(nele > 1)
+            os << "[";
+
+        bool done  = (nele == 0);
+        int idx = 0;
+
+        while(!done)
+        {
+            // if not first, add a comma prefix
+            if(idx > 0 )
+                os << ", ";
+
+            switch(m_dtype.id())
+            {
+                // ints
+                case DataType::INT8_ID:
+                case DataType::INT16_ID:
+                case DataType::INT32_ID:
+                case DataType::INT64_ID:
+                {
+                     os << (int64) element(idx);
+                     break;
+                }
+                // uints
+                case DataType::UINT8_ID:
+                case DataType::UINT16_ID:
+                case DataType::UINT32_ID:
+                case DataType::UINT64_ID:
+                {
+                    os << (uint64) element(idx);
+                    break;
+                }
+                // floats
+                case DataType::FLOAT32_ID:
+                case DataType::FLOAT64_ID:
+                {
+                    std::string fs = utils::float64_to_string((float64)element(idx));
+                    //check for inf and nan
+                    // looking for 'n' covers inf and nan
+                    bool inf_or_nan = fs.find('n') != std::string::npos;
+
+                    if(inf_or_nan)
+                        os << "\"";
+
+                    os << fs;
+
+                    if(inf_or_nan)
+                        os << "\"";
+                    break;
+                }
+                default:
+                {
+                    CONDUIT_ERROR("Leaf type \""
+                                  <<  m_dtype.name()
+                                  << "\""
+                                  << "is not supported in conduit::DataArray.")
+                }
+            }
+
+            idx++;
+
+            if(idx == bottom)
+            {
+                idx = nele - top;
+                os << ", ...";
+            }
+
+            if(idx == nele)
+            {
+                done = true;
+            }
+        }
+
+        if(nele > 1)
+            os << "]";
+    }
+}
+
+
+
 //-----------------------------------------------------------------------------
 //
 // -- conduit::DataArray explicit instantiations for supported array types --

diff --git a/src/libs/conduit/conduit_data_array.hpp b/src/libs/conduit/conduit_data_array.hpp
@@ -103,6 +103,14 @@ class CONDUIT_API DataArray
                                     Node &info,
                                     const float64 epsilon = CONDUIT_EPSILON) const;
 
+    ///
+    /// Summary Stats Helpers
+    ///
+    T               min()  const;
+    T               max()  const;
+    T               sum() const;
+    float64         mean() const;
+
 //-----------------------------------------------------------------------------
 // Setters
 //-----------------------------------------------------------------------------
@@ -207,7 +215,7 @@ class CONDUIT_API DataArray
     #endif
 
     #ifndef CONDUIT_USE_DOUBLE
-       void setconst std::initializer_list<double> &values);
+       void set(const std::initializer_list<double> &values);
     #endif
 
     //-------------------------------------------------------------------------
@@ -311,7 +319,14 @@ class CONDUIT_API DataArray
     void            to_yaml_stream(std::ostream &os) const;
 
     void            compact_elements_to(uint8 *data) const;
-
+
+    /// Creates a string repression for printing that limits
+    /// the number of elements shown to a max number
+    std::string     to_summary_string_default() const;
+    std::string     to_summary_string(index_t threshold=5) const;
+    void            to_summary_string_stream(std::ostream &os,
+                                             index_t threshold=5) const;
+
 //-----------------------------------------------------------------------------
 // -- stdout print methods ---
 //-----------------------------------------------------------------------------