polars-df 0.3.1-x86_64-linux → 0.5.0-x86_64-linux

data/README.md

@@ -50,14 +50,43 @@ From Parquet
 
 ```ruby
 Polars.read_parquet("file.parquet")
+
+# or lazily with
+Polars.scan_parquet("file.parquet")
 ```
 
 From Active Record
 
 ```ruby
-Polars.read_sql(User.all)
+Polars.read_database(User.all)
+# or
+Polars.read_database("SELECT * FROM users")
+```
+
+From JSON
+
+```ruby
+Polars.read_json("file.json")
 # or
-Polars.read_sql("SELECT * FROM users")
+Polars.read_ndjson("file.ndjson")
+
+# or lazily with
+Polars.scan_ndjson("file.ndjson")
+```
+
+From Feather / Arrow IPC
+
+```ruby
+Polars.read_ipc("file.arrow")
+
+# or lazily with
+Polars.scan_ipc("file.arrow")
+```
+
+From Avro
+
+```ruby
+Polars.read_avro("file.avro")
 ```
 
 From a hash

data/lib/polars/batched_csv_reader.rb

@@ -30,7 +30,7 @@ module Polars
       new_columns: nil
     )
       if Utils.pathlike?(file)
-        path = Utils.format_path(file)
+        path = Utils.normalise_filepath(file)
       end
 
       dtype_list = nil

data/lib/polars/binary_expr.rb

@@ -0,0 +1,77 @@
+module Polars
+  # Namespace for binary related expressions.
+  class BinaryExpr
+    # @private
+    attr_accessor :_rbexpr
+
+    # @private
+    def initialize(expr)
+      self._rbexpr = expr._rbexpr
+    end
+
+    # Check if binaries in Series contain a binary substring.
+    #
+    # @param lit [String]
+    #   The binary substring to look for
+    #
+    # @return [Expr]
+    def contains(lit)
+      Utils.wrap_expr(_rbexpr.binary_contains(lit))
+    end
+
+    # Check if string values end with a binary substring.
+    #
+    # @param sub [String]
+    #   Suffix substring.
+    #
+    # @return [Expr]
+    def ends_with(sub)
+      Utils.wrap_expr(_rbexpr.binary_ends_with(sub))
+    end
+
+    # Check if values start with a binary substring.
+    #
+    # @param sub [String]
+    #   Prefix substring.
+    #
+    # @return [Expr]
+    def starts_with(sub)
+      Utils.wrap_expr(_rbexpr.binary_starts_with(sub))
+    end
+
+    # Decode a value using the provided encoding.
+    #
+    # @param encoding ["hex", "base64"]
+    #   The encoding to use.
+    # @param strict [Boolean]
+    #   Raise an error if the underlying value cannot be decoded,
+    #   otherwise mask out with a null value.
+    #
+    # @return [Expr]
+    def decode(encoding, strict: true)
+      if encoding == "hex"
+        Utils.wrap_expr(_rbexpr.binary_hex_decode(strict))
+      elsif encoding == "base64"
+        Utils.wrap_expr(_rbexpr.binary_base64_decode(strict))
+      else
+        raise ArgumentError, "encoding must be one of {{'hex', 'base64'}}, got #{encoding}"
+      end
+    end
+
+    # Encode a value using the provided encoding.
+    #
+    # @param encoding ["hex", "base64"]
+    #   The encoding to use.
+    #
+    # @return [Expr]
+    def encode(encoding)
+      if encoding == "hex"
+        Utils.wrap_expr(_rbexpr.binary_hex_encode)
+      elsif encoding == "base64"
+        Utils.wrap_expr(_rbexpr.binary_base64_encode)
+      else
+        raise ArgumentError, "encoding must be one of {{'hex', 'base64'}}, got #{encoding}"
+      end
+    end
+  end
+end

data/lib/polars/binary_name_space.rb

@@ -0,0 +1,66 @@
+module Polars
+  # Series.bin namespace.
+  class BinaryNameSpace
+    include ExprDispatch
+
+    self._accessor = "bin"
+
+    # @private
+    def initialize(series)
+      self._s = series._s
+    end
+
+    # Check if binaries in Series contain a binary substring.
+    #
+    # @param lit [String]
+    #   The binary substring to look for
+    #
+    # @return [Series]
+    def contains(lit)
+      super
+    end
+
+    # Check if string values end with a binary substring.
+    #
+    # @param sub [String]
+    #   Suffix substring.
+    #
+    # @return [Series]
+    def ends_with(sub)
+      super
+    end
+
+    # Check if values start with a binary substring.
+    #
+    # @param sub [String]
+    #   Prefix substring.
+    #
+    # @return [Series]
+    def starts_with(sub)
+      super
+    end
+
+    # Decode a value using the provided encoding.
+    #
+    # @param encoding ["hex", "base64"]
+    #   The encoding to use.
+    # @param strict [Boolean]
+    #   Raise an error if the underlying value cannot be decoded,
+    #   otherwise mask out with a null value.
+    #
+    # @return [Series]
+    def decode(encoding, strict: true)
+      super
+    end
+
+    # Encode a value using the provided encoding.
+    #
+    # @param encoding ["hex", "base64"]
+    #   The encoding to use.
+    #
+    # @return [Series]
+    def encode(encoding)
+      super
+    end
+  end
+end

data/lib/polars/convert.rb

@@ -26,8 +26,8 @@ module Polars
     #   # │ 1   ┆ 3   │
     #   # │ 2   ┆ 4   │
     #   # └─────┴─────┘
-    def from_hash(data, columns: nil)
-      DataFrame._from_hash(data, columns: columns)
+    def from_hash(data, schema: nil, columns: nil)
+      DataFrame._from_hash(data, schema: schema || columns)
     end
 
     # Construct a DataFrame from a sequence of dictionaries. This operation clones data.