ruby-hdf5 0.0.1 → 0.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0a5d336f6b45f913a829873d20aba6c80c5724db14038965e073235f36477733
-  data.tar.gz: 1fe31c4c72258d31d57ca6ce043aee21f7cd9083700c77405a501a79f1ea5323
+  metadata.gz: 4591bafb2585ec00df81fc3831cc7037cc02dee81f15919ec7cf1b5d2f08a4c2
+  data.tar.gz: beb661f3c825adaab6b132cee69a402c1319d05143a0bec1516f2e0dec1180b2
 SHA512:
-  metadata.gz: 786821e23be9bdc249352c34dc5e293be4a6f3e931c69cc89a1f51ee284d53e00d2ccdb130ee49a2e5bf467ff89cf60f184c455ffc3271309ea0c37a189b28eb
-  data.tar.gz: 98150386a48675d3208416cca2c1d4f1668f9519c2acafbb2935ae0d8111789a1de890d1516640c2649bc463d8c07f1144acad050a018f209fe429cf5d394821
+  metadata.gz: bf13704291dfbdb2d0d61cf03504574a553e3d16e6db980355df5408ee425e340cfa1bc7b3850c4f877aa066958f20ab4cb5cbb07935f6302eb403771c6b4458
+  data.tar.gz: ac8511ab873fea84933c38c65c15c899cfb3dbc7a63d2a062f41d78722bcc87288d84997e5f296c38691deb2660e065046e7fdf13e2c611127ef8fc84b1ce2ea

data/README.md

@@ -11,14 +11,19 @@ This gem currently provides practical high-level wrappers for:
 - opening and creating files
 - creating groups
 - creating, writing, and reading one-dimensional numeric datasets
-- reading attributes
+- reading and writing numeric attributes
 
 Unsupported at this stage:
 
 - string dataset read/write
-- attribute write
 - multidimensional array write
 
+Integer datasets and integer attributes currently use native C `int` under the hood.
+Values outside that range are rejected with `HDF5::Error` to avoid silent overflow.
+
+`Group#list_datasets` filters datasets from group entries by checking object type per entry.
+For very large groups, this may be slower than `Group#list_entries`.
+
 ## Supported HDF5 Versions
 
 - HDF5 1.10

data/lib/hdf5/attribute.rb

@@ -3,18 +3,18 @@ module HDF5
     def initialize(dataset_id, attr_name)
       @dataset_id = dataset_id
       @attr_name = attr_name
-      @attr_id = FFI.H5Aopen(@dataset_id, @attr_name, 0)
-      raise 'Failed to open attribute' if @attr_id < 0
+      @attr_id = HDF5::FFI.H5Aopen(@dataset_id, @attr_name, HDF5::DEFAULT_PROPERTY_LIST)
+      raise HDF5::Error, 'Failed to open attribute' if @attr_id < 0
     end
 
     def read
-      type_id = FFI.H5Aget_type(@attr_id)
-      space_id = FFI.H5Aget_space(@attr_id)
+      type_id = HDF5::FFI.H5Aget_type(@attr_id)
+      space_id = HDF5::FFI.H5Aget_space(@attr_id)
 
-      size = FFI.H5Sget_simple_extent_npoints(space_id)
+      size = HDF5::FFI.H5Sget_simple_extent_npoints(space_id)
 
       buffer = \
-        case FFI.H5Tget_class(type_id)
+        case HDF5::FFI.H5Tget_class(type_id)
         when :H5T_INTEGER
           ::FFI::MemoryPointer.new(:int, size)
         when :H5T_FLOAT
@@ -22,13 +22,13 @@ module HDF5
         when :H5T_STRING
           ::FFI::MemoryPointer.new(:pointer, size)
         else
-          raise 'Unsupported data type'
+          raise HDF5::Error, 'Unsupported data type'
         end
 
-      status = FFI.H5Aread(@attr_id, type_id, buffer)
-      raise 'Failed to read attribute' if status < 0
+      status = HDF5::FFI.H5Aread(@attr_id, type_id, buffer)
+      raise HDF5::Error, 'Failed to read attribute' if status < 0
 
-      case FFI.H5Tget_class(type_id)
+      case HDF5::FFI.H5Tget_class(type_id)
       when :H5T_INTEGER
         buffer.read_array_of_int(size)
       when :H5T_FLOAT
@@ -36,12 +36,18 @@ module HDF5
       when :H5T_STRING
         buffer.read_pointer.read_string
       else
-        raise 'Unsupported data type'
+        raise HDF5::Error, 'Unsupported data type'
       end
+    ensure
+      HDF5::FFI.H5Tclose(type_id) if type_id && type_id >= 0
+      HDF5::FFI.H5Sclose(space_id) if space_id && space_id >= 0
     end
 
     def close
-      FFI.H5Aclose(@attr_id)
+      return if @attr_id.nil?
+
+      HDF5::FFI.H5Aclose(@attr_id)
+      @attr_id = nil
     end
   end
 
@@ -56,5 +62,95 @@ module HDF5
     ensure
       attr.close if attr
     end
+
+    def []=(attr_name, value)
+      write(attr_name, value)
+    end
+
+    def write(attr_name, value)
+      values = normalize_data(value)
+      datatype_id = datatype_id_for(values)
+
+      exists = HDF5::FFI.H5Aexists(@dataset_id, attr_name)
+      raise HDF5::Error, "Failed to check attribute existence: #{attr_name}" if exists.negative?
+
+      if exists.positive?
+        status = HDF5::FFI.H5Adelete(@dataset_id, attr_name)
+        raise HDF5::Error, "Failed to replace attribute: #{attr_name}" if status < 0
+      end
+
+      dims = ::FFI::MemoryPointer.new(:ulong_long, 1)
+      dims.write_array_of_ulong_long([values.length])
+      dataspace_id = HDF5::FFI.H5Screate_simple(1, dims, nil)
+      raise HDF5::Error, 'Failed to create attribute dataspace' if dataspace_id < 0
+
+      attr_id = HDF5::FFI.H5Acreate2(
+        @dataset_id,
+        attr_name,
+        datatype_id,
+        dataspace_id,
+        HDF5::DEFAULT_PROPERTY_LIST,
+        HDF5::DEFAULT_PROPERTY_LIST
+      )
+      raise HDF5::Error, "Failed to create attribute: #{attr_name}" if attr_id < 0
+
+      buffer = buffer_for(values)
+      status = HDF5::FFI.H5Awrite(attr_id, datatype_id, buffer)
+      raise HDF5::Error, "Failed to write attribute: #{attr_name}" if status < 0
+
+      value
+    ensure
+      HDF5::FFI.H5Aclose(attr_id) if attr_id && attr_id >= 0
+      HDF5::FFI.H5Sclose(dataspace_id) if dataspace_id && dataspace_id >= 0
+    end
+
+    private
+
+    def normalize_data(value)
+      values = value.is_a?(Array) ? value : [value]
+      raise HDF5::Error, 'Attribute data must not be empty' if values.empty?
+      raise HDF5::Error, 'Nested arrays are not supported for attributes' if values.any? { |item| item.is_a?(Array) }
+
+      values
+    end
+
+    def datatype_id_for(values)
+      if values.all? { |item| item.is_a?(Integer) }
+        validate_native_int_range!(values)
+        HDF5::FFI.H5T_NATIVE_INT
+      elsif values.all? { |item| item.is_a?(Numeric) }
+        HDF5::FFI.H5T_NATIVE_DOUBLE
+      else
+        raise HDF5::Error, 'Only numeric attribute data is supported'
+      end
+    end
+
+    def buffer_for(values)
+      if values.all? { |item| item.is_a?(Integer) }
+        buffer = ::FFI::MemoryPointer.new(:int, values.length)
+        buffer.write_array_of_int(values)
+      else
+        buffer = ::FFI::MemoryPointer.new(:double, values.length)
+        buffer.write_array_of_double(values.map(&:to_f))
+      end
+
+      buffer
+    end
+
+    def native_int_bounds
+      bits = ::FFI.type_size(:int) * 8
+      max = (1 << (bits - 1)) - 1
+      min = -(1 << (bits - 1))
+      [min, max]
+    end
+
+    def validate_native_int_range!(values)
+      min, max = native_int_bounds
+      out_of_range = values.find { |value| value < min || value > max }
+      return unless out_of_range
+
+      raise HDF5::Error,
+            "Integer value #{out_of_range} is outside native int range (#{min}..#{max}). Use a smaller value."
+    end
   end
 end

data/lib/hdf5/dataset.rb

@@ -1,11 +1,64 @@
 module HDF5
   class Dataset
+    module DataHelpers
+      module_function
+
+      def normalize_data(data)
+        values = data.is_a?(Array) ? data : [data]
+        raise HDF5::Error, 'Dataset data must not be empty' if values.empty?
+        raise HDF5::Error, 'Nested arrays are not supported' if values.any? { |value| value.is_a?(Array) }
+
+        values
+      end
+
+      def datatype_id_for(data)
+        if data.all? { |value| value.is_a?(Integer) }
+          validate_native_int_range!(data)
+          HDF5::FFI.H5T_NATIVE_INT
+        elsif data.all? { |value| value.is_a?(Numeric) }
+          HDF5::FFI.H5T_NATIVE_DOUBLE
+        else
+          raise HDF5::Error, 'Only numeric dataset data is supported'
+        end
+      end
+
+      def buffer_for(data)
+        if data.all? { |value| value.is_a?(Integer) }
+          buffer = ::FFI::MemoryPointer.new(:int, data.length)
+          buffer.write_array_of_int(data)
+        else
+          buffer = ::FFI::MemoryPointer.new(:double, data.length)
+          buffer.write_array_of_double(data.map(&:to_f))
+        end
+
+        buffer
+      end
+
+      def native_int_bounds
+        bits = ::FFI.type_size(:int) * 8
+        max = (1 << (bits - 1)) - 1
+        min = -(1 << (bits - 1))
+        [min, max]
+      end
+
+      def validate_native_int_range!(values)
+        min, max = native_int_bounds
+        out_of_range = values.find { |value| value < min || value > max }
+        return unless out_of_range
+
+        raise HDF5::Error,
+              "Integer value #{out_of_range} is outside native int range (#{min}..#{max}). Use a smaller value."
+      end
+    end
+
+    private_constant :DataHelpers
+
     class << self
       def create(parent_id, name, data)
-        values = normalize_data(data)
+        values = DataHelpers.normalize_data(data)
         dims = ::FFI::MemoryPointer.new(:ulong_long, 1)
         dims.write_array_of_ulong_long([values.length])
-        datatype_id = datatype_id_for(values)
+        datatype_id = DataHelpers.datatype_id_for(values)
         dataspace_id = HDF5::FFI.H5Screate_simple(1, dims, nil)
         raise HDF5::Error, "Failed to create dataspace for dataset: #{name}" if dataspace_id < 0
 
@@ -36,36 +89,6 @@ module HDF5
         end
       end
 
-      def normalize_data(data)
-        values = data.is_a?(Array) ? data : [data]
-        raise HDF5::Error, 'Dataset data must not be empty' if values.empty?
-        raise HDF5::Error, 'Nested arrays are not supported' if values.any? { |value| value.is_a?(Array) }
-
-        values
-      end
-
-      def datatype_id_for(data)
-        if data.all? { |value| value.is_a?(Integer) }
-          HDF5::FFI.H5T_NATIVE_INT
-        elsif data.all? { |value| value.is_a?(Numeric) }
-          HDF5::FFI.H5T_NATIVE_DOUBLE
-        else
-          raise HDF5::Error, 'Only numeric dataset data is supported'
-        end
-      end
-
-      def buffer_for(data)
-        if data.all? { |value| value.is_a?(Integer) }
-          buffer = ::FFI::MemoryPointer.new(:int, data.length)
-          buffer.write_array_of_int(data)
-        else
-          buffer = ::FFI::MemoryPointer.new(:double, data.length)
-          buffer.write_array_of_double(data.map(&:to_f))
-        end
-
-        buffer
-      end
-
       private
 
       def from_id(dataset_id, name)
@@ -84,9 +107,9 @@ module HDF5
     end
 
     def write(data)
-      values = self.class.normalize_data(data)
-      buffer = self.class.buffer_for(values)
-      mem_type_id = self.class.datatype_id_for(values)
+      values = DataHelpers.normalize_data(data)
+      mem_type_id = DataHelpers.datatype_id_for(values)
+      buffer = DataHelpers.buffer_for(values)
       status = HDF5::FFI.H5Dwrite(@dataset_id, mem_type_id, HDF5::DEFAULT_PROPERTY_LIST, HDF5::DEFAULT_PROPERTY_LIST,
                                   HDF5::DEFAULT_PROPERTY_LIST, buffer)
       raise HDF5::Error, 'Failed to write dataset' if status < 0
@@ -160,8 +183,8 @@ module HDF5
       buffer.read_array_of_double(total_elements)
     end
 
-    def read_string_data(total_elements)
-      raise NotImplementedError
+    def read_string_data(_total_elements)
+      raise HDF5::Error, 'String dataset reading is not supported yet'
     end
 
     private

data/lib/hdf5/file.rb

@@ -79,7 +79,7 @@ module HDF5
       elsif dataset?(name)
         Dataset.open(@file_id, name)
       else
-        raise HDF5::Error, 'Unknown object type'
+        raise HDF5::Error, "Group or dataset not found: #{name}"
       end
     end
 

data/lib/hdf5/group.rb

@@ -54,10 +54,10 @@ module HDF5
       Dataset.create(@group_id, name, data, &block)
     end
 
-    def list_datasets
-      datasets = []
+    def list_entries
+      entries = []
       callback = ::FFI::Function.new(:int, %i[int64_t string pointer pointer]) do |_, name, _, _|
-        datasets << name
+        entries << name
         0 # continue
       end
 
@@ -66,9 +66,13 @@ module HDF5
        else
          HDF5::FFI.H5Literate2(@group_id, :H5_INDEX_NAME, :H5_ITER_NATIVE, nil, callback, nil)
        end).negative? &&
-        raise('Failed to list datasets')
+        raise(HDF5::Error, 'Failed to list entries')
+
+      entries
+    end
 
-      datasets
+    def list_datasets
+      list_entries.select { |name| dataset?(name) }
     end
 
     def [](name)
@@ -77,7 +81,7 @@ module HDF5
       elsif dataset?(name)
         Dataset.open(@group_id, name)
       else
-        raise HDF5::Error, 'Group or Dataset not found'
+        raise HDF5::Error, "Group or dataset not found: #{name}"
       end
     end
 

data/lib/hdf5/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module HDF5
-  VERSION = '0.0.1'
+  VERSION = '0.0.2'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: ruby-hdf5
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
 platform: ruby
 authors:
 - kojix2