polars-df 0.13.0 → 0.14.0

data/ext/polars/src/series/construction.rs

@@ -1,4 +1,4 @@
-use magnus::{prelude::*, RArray};
+use magnus::{prelude::*, RArray, RString};
 use polars_core::prelude::*;
 
 use crate::any_value::rb_object_to_any_value;
@@ -10,7 +10,7 @@ use crate::{RbPolarsErr, RbResult, RbSeries, RbTypeError, RbValueError};
 impl RbSeries {
     pub fn new_opt_bool(name: String, obj: RArray, strict: bool) -> RbResult<RbSeries> {
         let len = obj.len();
-        let mut builder = BooleanChunkedBuilder::new(&name, len);
+        let mut builder = BooleanChunkedBuilder::new(name.into(), len);
 
         unsafe {
             for item in obj.as_slice().iter() {
@@ -43,7 +43,7 @@ where
     T::Native: magnus::TryConvert,
 {
     let len = values.len();
-    let mut builder = PrimitiveChunkedBuilder::<T>::new(name, len);
+    let mut builder = PrimitiveChunkedBuilder::<T>::new(name.into(), len);
 
     for res in values.into_iter() {
         let value = res;
@@ -96,7 +96,7 @@ impl RbSeries {
         // from anyvalues is fallible
         let result = any_values_result.and_then(|avs| {
             let avs = slice_extract_wrapped(&avs);
-            let s = Series::from_any_values(&name, avs, strict).map_err(|e| {
+            let s = Series::from_any_values(name.clone().into(), avs, strict).map_err(|e| {
                 RbTypeError::new_err(format!(
                     "{e}\n\nHint: Try setting `strict: false` to allow passing data with mixed types."
                 ))
@@ -122,44 +122,71 @@ impl RbSeries {
             .into_iter()
             .map(|v| rb_object_to_any_value(v, strict))
             .collect::<RbResult<Vec<AnyValue>>>()?;
-        let s = Series::from_any_values_and_dtype(&name, any_values.as_slice(), &dtype.0, strict)
-            .map_err(|e| {
-            RbTypeError::new_err(format!(
+        let s =
+            Series::from_any_values_and_dtype(name.into(), any_values.as_slice(), &dtype.0, strict)
+                .map_err(|e| {
+                    RbTypeError::new_err(format!(
                 "{e}\n\nHint: Try setting `strict: false` to allow passing data with mixed types."
             ))
-        })?;
+                })?;
         Ok(s.into())
     }
 
-    pub fn new_str(name: String, val: Wrap<StringChunked>, _strict: bool) -> Self {
-        let mut s = val.0.into_series();
-        s.rename(&name);
-        RbSeries::new(s)
-    }
+    pub fn new_str(name: String, values: RArray, _strict: bool) -> RbResult<Self> {
+        let len = values.len();
+        let mut builder = StringChunkedBuilder::new(name.into(), len);
+
+        for res in values.into_iter() {
+            let value = res;
+            if value.is_nil() {
+                builder.append_null()
+            } else {
+                let v = String::try_convert(value)?;
+                builder.append_value(v)
+            }
+        }
 
-    pub fn new_binary(name: String, val: Wrap<BinaryChunked>, _strict: bool) -> Self {
-        let mut s = val.0.into_series();
-        s.rename(&name);
-        RbSeries::new(s)
+        let ca = builder.finish();
+        let s = ca.into_series();
+        Ok(s.into())
     }
 
-    pub fn new_null(name: String, val: RArray, _strict: bool) -> RbResult<Self> {
-        let s = Series::new_null(&name, val.len());
+    pub fn new_binary(name: String, values: RArray, _strict: bool) -> RbResult<Self> {
+        let len = values.len();
+        let mut builder = BinaryChunkedBuilder::new(name.into(), len);
+
+        for res in values.into_iter() {
+            let value = res;
+            if value.is_nil() {
+                builder.append_null()
+            } else {
+                let v = RString::try_convert(value)?;
+                builder.append_value(unsafe { v.as_slice() })
+            }
+        }
+
+        let ca = builder.finish();
+        let s = ca.into_series();
         Ok(s.into())
     }
 
+    pub fn new_null(name: String, values: RArray, _strict: bool) -> RbResult<Self> {
+        let len = values.len();
+        Ok(Series::new_null(name.into(), len).into())
+    }
+
     pub fn new_object(name: String, val: RArray, _strict: bool) -> RbResult<Self> {
         let val = val
             .into_iter()
             .map(ObjectValue::from)
             .collect::<Vec<ObjectValue>>();
-        let s = ObjectChunked::<ObjectValue>::new_from_vec(&name, val).into_series();
+        let s = ObjectChunked::<ObjectValue>::new_from_vec(name.into(), val).into_series();
         Ok(s.into())
     }
 
     pub fn new_series_list(name: String, val: RArray, _strict: bool) -> RbResult<Self> {
         let series_vec = to_series_collection(val)?;
-        Ok(Series::new(&name, &series_vec).into())
+        Ok(Series::new(name.into(), &series_vec).into())
     }
 
     pub fn new_array(
@@ -171,7 +198,7 @@ impl RbSeries {
     ) -> RbResult<Self> {
         let val = vec_wrap_any_value(val)?;
         let val = vec_extract_wrapped(val);
-        let out = Series::new(&name, &val);
+        let out = Series::new(name.into(), &val);
         match out.dtype() {
             DataType::List(list_inner) => {
                 let out = out
@@ -199,7 +226,7 @@ impl RbSeries {
         dtype: Wrap<DataType>,
     ) -> RbResult<Self> {
         let av = val.0;
-        Ok(Series::new(&name, &[av])
+        Ok(Series::new(name.into(), &[av])
             .cast(&dtype.0)
             .map_err(RbPolarsErr::from)?
             .new_from_index(0, n)

data/ext/polars/src/series/mod.rs

@@ -161,11 +161,11 @@ impl RbSeries {
     }
 
     pub fn name(&self) -> String {
-        self.series.borrow().name().into()
+        self.series.borrow().name().to_string()
     }
 
     pub fn rename(&self, name: String) {
-        self.series.borrow_mut().rename(&name);
+        self.series.borrow_mut().rename(name.into());
     }
 
     pub fn dtype(&self) -> Value {
@@ -257,7 +257,7 @@ impl RbSeries {
         let out = self
             .series
             .borrow()
-            .value_counts(sort, parallel, name, normalize)
+            .value_counts(sort, parallel, name.into(), normalize)
             .map_err(RbPolarsErr::from)?;
         Ok(out.into())
     }
@@ -395,7 +395,7 @@ impl RbSeries {
                     .0
             });
             avs.extend(iter);
-            return Ok(Series::new(&self.name(), &avs).into());
+            return Ok(Series::new(self.name().into(), &avs).into());
         }
 
         let out = match output_type {

data/lib/polars/data_frame.rb

@@ -2426,15 +2426,15 @@ module Polars
     #   df.map_rows { |t| t[0] * 2 + t[1] }
     #   # =>
     #   # shape: (3, 1)
-    #   # ┌───────┐
-    #   # │ apply │
-    #   # │ ---   │
-    #   # │ i64   │
-    #   # ╞═══════╡
-    #   # │ 1     │
-    #   # │ 9     │
-    #   # │ 14    │
-    #   # └───────┘
+    #   # ┌─────┐
+    #   # │ map │
+    #   # │ --- │
+    #   # │ i64 │
+    #   # ╞═════╡
+    #   # │ 1   │
+    #   # │ 9   │
+    #   # │ 14  │
+    #   # └─────┘
     def map_rows(return_dtype: nil, inference_size: 256, &f)
       out, is_df = _df.map_rows(f, return_dtype, inference_size)
       if is_df
@@ -4234,7 +4234,7 @@ module Polars
       if n.nil? && !frac.nil?
         frac = Series.new("frac", [frac]) unless frac.is_a?(Series)
 
-        _from_rbdf(
+        return _from_rbdf(
           _df.sample_frac(frac._s, with_replacement, shuffle, seed)
         )
       end

data/lib/polars/expr.rb

@@ -1182,7 +1182,7 @@ module Polars
     #       "b" => [1, 1, 2, 2]
     #     }
     #   )
-    #   df.select(Polars.all.mode)
+    #   df.select(Polars.all.mode.first)
     #   # =>
     #   # shape: (2, 2)
     #   # ┌─────┬─────┐
@@ -6015,12 +6015,12 @@ module Polars
     #   # ┌──────┐
     #   # │ a    │
     #   # │ ---  │
-    #   # │ i64  │
+    #   # │ f64  │
     #   # ╞══════╡
-    #   # │ -1   │
-    #   # │ 0    │
-    #   # │ 0    │
-    #   # │ 1    │
+    #   # │ -1.0 │
+    #   # │ -0.0 │
+    #   # │ 0.0  │
+    #   # │ 1.0  │
     #   # │ null │
     #   # └──────┘
     def sign

data/lib/polars/io/ipc.rb

@@ -189,10 +189,6 @@ module Polars
     #   Offset to start the row_count column (only use if the name is set).
     # @param storage_options [Hash]
     #   Extra options that make sense for a particular storage connection.
-    # @param memory_map [Boolean]
-    #   Try to memory map the file. This can greatly improve performance on repeated
-    #   queries as the OS may cache pages.
-    #   Only uncompressed IPC files can be memory mapped.
     # @param hive_partitioning [Boolean]
     #   Infer statistics and schema from Hive partitioned URL and use them
     #   to prune reads. This is unset by default (i.e. `nil`), meaning it is
@@ -215,7 +211,6 @@ module Polars
       row_count_name: nil,
       row_count_offset: 0,
       storage_options: nil,
-      memory_map: true,
       hive_partitioning: nil,
       hive_schema: nil,
       try_parse_hive_dates: true,
@@ -229,7 +224,6 @@ module Polars
         row_count_name: row_count_name,
         row_count_offset: row_count_offset,
         storage_options: storage_options,
-        memory_map: memory_map,
         hive_partitioning: hive_partitioning,
         hive_schema: hive_schema,
         try_parse_hive_dates: try_parse_hive_dates,
@@ -246,7 +240,6 @@ module Polars
       row_count_name: nil,
       row_count_offset: 0,
       storage_options: nil,
-      memory_map: true,
       hive_partitioning: nil,
       hive_schema: nil,
       try_parse_hive_dates: true,
@@ -263,7 +256,6 @@ module Polars
           cache,
           rechunk,
           Utils.parse_row_index_args(row_count_name, row_count_offset),
-          memory_map,
           hive_partitioning,
           hive_schema,
           try_parse_hive_dates,

data/lib/polars/series.rb

@@ -2606,12 +2606,12 @@ module Polars
     #   s.sign
     #   # =>
     #   # shape: (5,)
-    #   # Series: 'a' [i64]
+    #   # Series: 'a' [f64]
     #   # [
-    #   #         -1
-    #   #         0
-    #   #         0
-    #   #         1
+    #   #         -1.0
+    #   #         -0.0
+    #   #         0.0
+    #   #         1.0
     #   #         null
     #   # ]
     def sign

data/lib/polars/version.rb

@@ -1,4 +1,4 @@
 module Polars
   # @private
-  VERSION = "0.13.0"
+  VERSION = "0.14.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: polars-df
 version: !ruby/object:Gem::Version
-  version: 0.13.0
+  version: 0.14.0
 platform: ruby
 authors:
 - Andrew Kane
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-09-05 00:00:00.000000000 Z
+date: 2024-09-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bigdecimal
@@ -199,7 +199,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.5.11
+rubygems_version: 3.5.16
 signing_key:
 specification_version: 4
 summary: Blazingly fast DataFrames for Ruby