polars-df 0.3.1-x86_64-linux → 0.4.0-x86_64-linux

data/lib/polars/io.rb

@@ -2,7 +2,7 @@ module Polars
   module IO
     # Read a CSV file into a DataFrame.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     # @param has_header [Boolean]
     #   Indicate if the first row of dataset is a header or not.
@@ -89,7 +89,7 @@ module Polars
     #   Set `rechunk: false` if you are benchmarking the csv-reader. A `rechunk` is
     #   an expensive operation.
     def read_csv(
-      file,
+      source,
       has_header: true,
       columns: nil,
       new_columns: nil,
@@ -137,7 +137,7 @@ module Polars
       end
 
       df = nil
-      _prepare_file_arg(file) do |data|
+      _prepare_file_arg(source) do |data|
         df = DataFrame._read_csv(
           data,
           has_header: has_header,
@@ -178,7 +178,7 @@ module Polars
     # projections to the scan level, thereby potentially reducing
     # memory overhead.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file.
     # @param has_header [Boolean]
     #   Indicate if the first row of dataset is a header or not.
@@ -242,7 +242,7 @@ module Polars
     #
     # @return [LazyFrame]
     def scan_csv(
-      file,
+      source,
       has_header: true,
       sep: ",",
       comment_char: nil,
@@ -268,12 +268,12 @@ module Polars
       _check_arg_is_1byte("comment_char", comment_char, false)
       _check_arg_is_1byte("quote_char", quote_char, true)
 
-      if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+      if Utils.pathlike?(source)
+        source = Utils.normalise_filepath(source)
       end
 
       LazyFrame._scan_csv(
-        file,
+        source,
         has_header: has_header,
         sep: sep,
         comment_char: comment_char,
@@ -302,7 +302,7 @@ module Polars
     # This allows the query optimizer to push down predicates and projections to the scan
     # level, thereby potentially reducing memory overhead.
     #
-    # @param file [String]
+    # @param source [String]
     #   Path to a IPC file.
     # @param n_rows [Integer]
     #   Stop reading from IPC file after reading `n_rows`.
@@ -324,7 +324,7 @@ module Polars
     #
     # @return [LazyFrame]
     def scan_ipc(
-      file,
+      source,
       n_rows: nil,
       cache: true,
       rechunk: true,
@@ -334,7 +334,7 @@ module Polars
       memory_map: true
     )
       LazyFrame._scan_ipc(
-        file,
+        source,
         n_rows: n_rows,
         cache: cache,
         rechunk: rechunk,
@@ -350,7 +350,7 @@ module Polars
     # This allows the query optimizer to push down predicates and projections to the scan
     # level, thereby potentially reducing memory overhead.
     #
-    # @param file [String]
+    # @param source [String]
     #   Path to a file.
     # @param n_rows [Integer]
     #   Stop reading from parquet file after reading `n_rows`.
@@ -374,7 +374,7 @@ module Polars
     #
     # @return [LazyFrame]
     def scan_parquet(
-      file,
+      source,
       n_rows: nil,
       cache: true,
       parallel: "auto",
@@ -384,12 +384,12 @@ module Polars
       storage_options: nil,
       low_memory: false
     )
-      if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+      if Utils.pathlike?(source)
+        source = Utils.normalise_filepath(source)
       end
 
       LazyFrame._scan_parquet(
-        file,
+        source,
         n_rows:n_rows,
         cache: cache,
         parallel: parallel,
@@ -406,7 +406,7 @@ module Polars
     # This allows the query optimizer to push down predicates and projections to the scan
     # level, thereby potentially reducing memory overhead.
     #
-    # @param file [String]
+    # @param source [String]
     #   Path to a file.
     # @param infer_schema_length [Integer]
     #   Infer the schema length from the first `infer_schema_length` rows.
@@ -426,7 +426,7 @@ module Polars
     #
     # @return [LazyFrame]
     def scan_ndjson(
-      file,
+      source,
       infer_schema_length: 100,
       batch_size: 1024,
       n_rows: nil,
@@ -435,12 +435,12 @@ module Polars
       row_count_name: nil,
       row_count_offset: 0
     )
-      if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+      if Utils.pathlike?(source)
+        source = Utils.normalise_filepath(source)
       end
 
       LazyFrame._scan_ndjson(
-        file,
+        source,
         infer_schema_length: infer_schema_length,
         batch_size: batch_size,
         n_rows: n_rows,
@@ -453,7 +453,7 @@ module Polars
 
     # Read into a DataFrame from Apache Avro format.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     # @param columns [Object]
     #   Columns to select. Accepts a list of column indices (starting at zero) or a list
@@ -462,17 +462,17 @@ module Polars
     #   Stop reading from Apache Avro file after reading ``n_rows``.
     #
     # @return [DataFrame]
-    def read_avro(file, columns: nil, n_rows: nil)
-      if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+    def read_avro(source, columns: nil, n_rows: nil)
+      if Utils.pathlike?(source)
+        source = Utils.normalise_filepath(source)
       end
 
-      DataFrame._read_avro(file, n_rows: n_rows, columns: columns)
+      DataFrame._read_avro(source, n_rows: n_rows, columns: columns)
     end
 
     # Read into a DataFrame from Arrow IPC (Feather v2) file.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     # @param columns [Object]
     #   Columns to select. Accepts a list of column indices (starting at zero) or a list
@@ -495,7 +495,7 @@ module Polars
     #
     # @return [DataFrame]
     def read_ipc(
-      file,
+      source,
       columns: nil,
       n_rows: nil,
       memory_map: true,
@@ -505,7 +505,7 @@ module Polars
       rechunk: true
     )
       storage_options ||= {}
-      _prepare_file_arg(file, **storage_options) do |data|
+      _prepare_file_arg(source, **storage_options) do |data|
         DataFrame._read_ipc(
           data,
           columns: columns,
@@ -520,8 +520,8 @@ module Polars
 
     # Read into a DataFrame from a parquet file.
     #
-    # @param file [Object]
-    #   Path to a file, or a file-like object.
+    # @param source [Object]
+    #   Path to a file or a file-like object.
     # @param columns [Object]
     #   Columns to select. Accepts a list of column indices (starting at zero) or a list
     #   of column names.
@@ -539,6 +539,12 @@ module Polars
     #   Offset to start the row_count column (only use if the name is set).
     # @param low_memory [Boolean]
     #   Reduce memory pressure at the expense of performance.
+    # @param use_statistics [Boolean]
+    #   Use statistics in the parquet to determine if pages
+    #   can be skipped from reading.
+    # @param rechunk [Boolean]
+    #   Make sure that all columns are contiguous in memory by
+    #   aggregating the chunks into a single array.
     #
     # @return [DataFrame]
     #
@@ -548,16 +554,18 @@ module Polars
     #   Set `rechunk: false` if you are benchmarking the parquet-reader. A `rechunk` is
     #   an expensive operation.
     def read_parquet(
-      file,
+      source,
       columns: nil,
       n_rows: nil,
       storage_options: nil,
       parallel: "auto",
       row_count_name: nil,
       row_count_offset: 0,
-      low_memory: false
+      low_memory: false,
+      use_statistics: true,
+      rechunk: true
     )
-      _prepare_file_arg(file) do |data|
+      _prepare_file_arg(source) do |data|
         DataFrame._read_parquet(
           data,
           columns: columns,
@@ -565,49 +573,51 @@ module Polars
           parallel: parallel,
           row_count_name: row_count_name,
           row_count_offset: row_count_offset,
-          low_memory: low_memory
+          low_memory: low_memory,
+          use_statistics: use_statistics,
+          rechunk: rechunk
         )
       end
     end
 
     # Read into a DataFrame from a JSON file.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     #
     # @return [DataFrame]
-    def read_json(file)
-      DataFrame._read_json(file)
+    def read_json(source)
+      DataFrame._read_json(source)
     end
 
     # Read into a DataFrame from a newline delimited JSON file.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     #
     # @return [DataFrame]
-    def read_ndjson(file)
-      DataFrame._read_ndjson(file)
+    def read_ndjson(source)
+      DataFrame._read_ndjson(source)
     end
 
     # Read a SQL query into a DataFrame.
     #
-    # @param sql [Object]
+    # @param query [Object]
     #   ActiveRecord::Relation or ActiveRecord::Result.
     #
     # @return [DataFrame]
-    def read_sql(sql)
+    def read_database(query)
       if !defined?(ActiveRecord)
         raise Error, "Active Record not available"
       end
 
       result =
-        if sql.is_a?(ActiveRecord::Result)
-          sql
-        elsif sql.is_a?(ActiveRecord::Relation)
-          sql.connection.select_all(sql.to_sql)
-        elsif sql.is_a?(String)
-          ActiveRecord::Base.connection.select_all(sql)
+        if query.is_a?(ActiveRecord::Result)
+          query
+        elsif query.is_a?(ActiveRecord::Relation)
+          query.connection.select_all(query.to_sql)
+        elsif query.is_a?(String)
+          ActiveRecord::Base.connection.select_all(query)
         else
           raise ArgumentError, "Expected ActiveRecord::Relation, ActiveRecord::Result, or String"
         end
@@ -617,6 +627,7 @@ module Polars
       end
       DataFrame.new(data)
     end
+    alias_method :read_sql, :read_database
 
     # def read_excel
     # end
@@ -628,7 +639,7 @@ module Polars
     # file chunks. After that work will only be done
     # if `next_batches` is called.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     # @param has_header [Boolean]
     #   Indicate if the first row of dataset is a header or not.
@@ -712,7 +723,7 @@ module Polars
     #   )
     #   reader.next_batches(5)
     def read_csv_batched(
-      file,
+      source,
       has_header: true,
       columns: nil,
       new_columns: nil,
@@ -752,7 +763,7 @@ module Polars
       end
 
       BatchedCsvReader.new(
-        file,
+        source,
         has_header: has_header,
         columns: columns || projection,
         sep: sep,
@@ -781,30 +792,30 @@ module Polars
 
     # Get a schema of the IPC file without reading data.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     #
     # @return [Hash]
-    def read_ipc_schema(file)
-      if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+    def read_ipc_schema(source)
+      if Utils.pathlike?(source)
+        source = Utils.normalise_filepath(source)
       end
 
-      _ipc_schema(file)
+      _ipc_schema(source)
     end
 
     # Get a schema of the Parquet file without reading data.
     #
-    # @param file [Object]
+    # @param source [Object]
     #   Path to a file or a file-like object.
     #
     # @return [Hash]
-    def read_parquet_schema(file)
-      if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+    def read_parquet_schema(source)
+      if Utils.pathlike?(source)
+        source = Utils.normalise_filepath(source)
       end
 
-      _parquet_schema(file)
+      _parquet_schema(source)
     end
 
     private

data/lib/polars/lazy_frame.rb

@@ -80,7 +80,8 @@ module Polars
       row_count_name: nil,
       row_count_offset: 0,
       storage_options: nil,
-      low_memory: false
+      low_memory: false,
+      use_statistics: true
     )
       _from_rbldf(
         RbLazyFrame.new_from_parquet(
@@ -90,7 +91,8 @@ module Polars
           parallel,
           rechunk,
           Utils._prepare_row_count_args(row_count_name, row_count_offset),
-          low_memory
+          low_memory,
+          use_statistics
         )
       )
     end
@@ -107,7 +109,7 @@ module Polars
       memory_map: true
     )
       if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+        file = Utils.normalise_filepath(file)
       end
 
       _from_rbldf(
@@ -157,7 +159,7 @@ module Polars
     # @return [LazyFrame]
     def self.read_json(file)
       if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+        file = Utils.normalise_filepath(file)
       end
 
       Utils.wrap_ldf(RbLazyFrame.read_json(file))
@@ -264,7 +266,7 @@ module Polars
     # @return [nil]
     def write_json(file)
       if Utils.pathlike?(file)
-        file = Utils.format_path(file)
+        file = Utils.normalise_filepath(file)
       end
       _ldf.write_json(file)
       nil
@@ -473,6 +475,96 @@ module Polars
       Utils.wrap_df(ldf.collect)
     end
 
+    # Persists a LazyFrame at the provided path.
+    #
+    # This allows streaming results that are larger than RAM to be written to disk.
+    #
+    # @param path [String]
+    #   File path to which the file should be written.
+    # @param compression ["lz4", "uncompressed", "snappy", "gzip", "lzo", "brotli", "zstd"]
+    #   Choose "zstd" for good compression performance.
+    #   Choose "lz4" for fast compression/decompression.
+    #   Choose "snappy" for more backwards compatibility guarantees
+    #   when you deal with older parquet readers.
+    # @param compression_level [Integer]
+    #   The level of compression to use. Higher compression means smaller files on
+    #   disk.
+    #
+    #   - "gzip" : min-level: 0, max-level: 10.
+    #   - "brotli" : min-level: 0, max-level: 11.
+    #   - "zstd" : min-level: 1, max-level: 22.
+    # @param statistics [Boolean]
+    #   Write statistics to the parquet headers. This requires extra compute.
+    # @param row_group_size [Integer]
+    #   Size of the row groups in number of rows.
+    #   If `nil` (default), the chunks of the `DataFrame` are
+    #   used. Writing in smaller chunks may reduce memory pressure and improve
+    #   writing speeds.
+    # @param data_pagesize_limit [Integer]
+    #   Size limit of individual data pages.
+    #   If not set defaults to 1024 * 1024 bytes
+    # @param maintain_order [Boolean]
+    #   Maintain the order in which data is processed.
+    #   Setting this to `false` will  be slightly faster.
+    # @param type_coercion [Boolean]
+    #   Do type coercion optimization.
+    # @param predicate_pushdown [Boolean]
+    #   Do predicate pushdown optimization.
+    # @param projection_pushdown [Boolean]
+    #   Do projection pushdown optimization.
+    # @param simplify_expression [Boolean]
+    #   Run simplify expressions optimization.
+    # @param no_optimization [Boolean]
+    #   Turn off (certain) optimizations.
+    # @param slice_pushdown [Boolean]
+    #   Slice pushdown optimization.
+    #
+    # @return [DataFrame]
+    #
+    # @example
+    #   lf = Polars.scan_csv("/path/to/my_larger_than_ram_file.csv")
+    #   lf.sink_parquet("out.parquet")
+    def sink_parquet(
+      path,
+      compression: "zstd",
+      compression_level: nil,
+      statistics: false,
+      row_group_size: nil,
+      data_pagesize_limit: nil,
+      maintain_order: true,
+      type_coercion: true,
+      predicate_pushdown: true,
+      projection_pushdown: true,
+      simplify_expression: true,
+      no_optimization: false,
+      slice_pushdown: true
+    )
+      if no_optimization
+        predicate_pushdown = false
+        projection_pushdown = false
+        slice_pushdown = false
+      end
+
+      lf = _ldf.optimization_toggle(
+        type_coercion,
+        predicate_pushdown,
+        projection_pushdown,
+        simplify_expression,
+        slice_pushdown,
+        false,
+        true
+      )
+      lf.sink_parquet(
+        path,
+        compression,
+        compression_level,
+        statistics,
+        row_group_size,
+        data_pagesize_limit,
+        maintain_order
+      )
+    end
+
     # Collect a small number of rows for debugging purposes.
     #
     # Fetch is like a {#collect} operation, but it overwrites the number of rows
@@ -2192,6 +2284,10 @@ module Polars
     #   Name to give to the `value` column. Defaults to "variable"
     # @param value_name [String]
     #   Name to give to the `value` column. Defaults to "value"
+    # @param streamable [Boolean]
+    #   Allow this node to run in the streaming engine.
+    #   If this runs in streaming, the output of the melt operation
+    #   will not have a stable ordering.
     #
     # @return [LazyFrame]
     #
@@ -2218,7 +2314,7 @@ module Polars
     #   # │ y   ┆ c        ┆ 4     │
     #   # │ z   ┆ c        ┆ 6     │
     #   # └─────┴──────────┴───────┘
-    def melt(id_vars: nil, value_vars: nil, variable_name: nil, value_name: nil)
+    def melt(id_vars: nil, value_vars: nil, variable_name: nil, value_name: nil, streamable: true)
       if value_vars.is_a?(String)
         value_vars = [value_vars]
       end
@@ -2232,7 +2328,7 @@ module Polars
         id_vars = []
       end
       _from_rbldf(
-        _ldf.melt(id_vars, value_vars, value_name, variable_name)
+        _ldf.melt(id_vars, value_vars, value_name, variable_name, streamable)
       )
     end
 

data/lib/polars/lazy_functions.rb

@@ -657,7 +657,7 @@ module Polars
     #   Default is ascending.
     #
     # @return [Expr]
-    def argsort_by(exprs, reverse: false)
+    def arg_sort_by(exprs, reverse: false)
       if !exprs.is_a?(Array)
         exprs = [exprs]
       end
@@ -665,8 +665,9 @@ module Polars
         reverse = [reverse] * exprs.length
       end
       exprs = Utils.selection_to_rbexpr_list(exprs)
-      Utils.wrap_expr(RbExpr.argsort_by(exprs, reverse))
+      Utils.wrap_expr(RbExpr.arg_sort_by(exprs, reverse))
     end
+    alias_method :argsort_by, :arg_sort_by
 
     # Create polars `Duration` from distinct time components.
     #

data/lib/polars/list_expr.rb

@@ -426,7 +426,7 @@ module Polars
     #   # shape: (2,)
     #   # Series: 'a' [list[i64]]
     #   # [
-    #   #         [null, 1, ... 1]
+    #   #         [null, 1, … 1]
     #   #         [null, -8, -1]
     #   # ]
     def diff(n: 1, null_behavior: "ignore")
@@ -447,7 +447,7 @@ module Polars
     #   # shape: (2,)
     #   # Series: 'a' [list[i64]]
     #   # [
-    #   #         [null, 1, ... 3]
+    #   #         [null, 1, … 3]
     #   #         [null, 10, 2]
     #   # ]
     def shift(periods = 1)

data/lib/polars/list_name_space.rb

@@ -185,7 +185,7 @@ module Polars
     #   # shape: (2,)
     #   # Series: 'a' [list[i64]]
     #   # [
-    #   #         [null, 1, ... 1]
+    #   #         [null, 1, … 1]
     #   #         [null, -8, -1]
     #   # ]
     def diff(n: 1, null_behavior: "ignore")
@@ -206,7 +206,7 @@ module Polars
     #   # shape: (2,)
     #   # Series: 'a' [list[i64]]
     #   # [
-    #   #         [null, 1, ... 3]
+    #   #         [null, 1, … 3]
     #   #         [null, 10, 2]
     #   # ]
     def shift(periods = 1)

data/lib/polars/series.rb

@@ -3531,6 +3531,13 @@ module Polars
       ListNameSpace.new(self)
     end
 
+    # Create an object namespace of all binary related methods.
+    #
+    # @return [BinaryNameSpace]
+    def bin
+      BinaryNameSpace.new(self)
+    end
+
     # Create an object namespace of all categorical related methods.
     #
     # @return [CatNameSpace]
@@ -3795,7 +3802,8 @@ module Polars
       UInt32 => RbSeries.method(:new_opt_u32),
       UInt64 => RbSeries.method(:new_opt_u64),
       Boolean => RbSeries.method(:new_opt_bool),
-      Utf8 => RbSeries.method(:new_str)
+      Utf8 => RbSeries.method(:new_str),
+      Binary => RbSeries.method(:new_binary)
     }
 
     SYM_TYPE_TO_CONSTRUCTOR = {

data/lib/polars/string_expr.rb

@@ -332,7 +332,7 @@ module Polars
     #   # │ -0001   │
     #   # │ 00000   │
     #   # │ 00001   │
-    #   # │ ...     │
+    #   # │ …       │
     #   # │ 10000   │
     #   # │ 100000  │
     #   # │ 1000000 │

data/lib/polars/utils.rb

@@ -93,8 +93,12 @@ module Polars
       Polars.lit(value)
     end
 
-    def self.format_path(path)
-      File.expand_path(path)
+    def self.normalise_filepath(path, check_not_directory: true)
+      path = File.expand_path(path)
+      if check_not_directory && File.exist?(path) && Dir.exist?(path)
+        raise ArgumentError, "Expected a file path; #{path} is a directory"
+      end
+      path
     end
 
     # TODO fix
@@ -216,5 +220,9 @@ module Polars
         val.is_a?(Array) && _is_iterable_of(val, String)
       end
     end
+
+    def self.local_file?(file)
+      Dir.glob(file).any?
+    end
   end
 end

data/lib/polars/version.rb

@@ -1,4 +1,4 @@
 module Polars
   # @private
-  VERSION = "0.3.1"
+  VERSION = "0.4.0"
 end

data/lib/polars.rb

@@ -12,6 +12,8 @@ require "stringio"
 # modules
 require_relative "polars/expr_dispatch"
 require_relative "polars/batched_csv_reader"
+require_relative "polars/binary_expr"
+require_relative "polars/binary_name_space"
 require_relative "polars/cat_expr"
 require_relative "polars/cat_name_space"
 require_relative "polars/convert"