From 2ead45e1f827fbe91ba1bf4ed699e0344ffa4e93 Mon Sep 17 00:00:00 2001
From: wangfeng <wangfengxy@sina.cn>
Date: Wed, 19 Apr 2023 11:43:16 +0800
Subject: [PATCH 1/3] =?UTF-8?q?dataframe=E6=94=AF=E6=8C=81=E4=BB=8E?=
 =?UTF-8?q?=E5=90=8E=E5=88=B0=E5=89=8D=E7=9A=84=E9=80=89=E6=8B=A9=E4=B8=80?=
 =?UTF-8?q?=E4=B8=AA=E5=AD=90=E9=9B=86?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 dataframe_subset.go | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/dataframe_subset.go b/dataframe_subset.go
index 747f277..b62ea09 100644
--- a/dataframe_subset.go
+++ b/dataframe_subset.go
@@ -26,7 +26,13 @@ func (self DataFrame) Subset(start, end int) DataFrame {
 	}
 }
 
-// Select 选择一段记录
+// Sub 选择一个子集, start end 支持从后到前选择
+func (self DataFrame) Sub(start, end int) DataFrame {
+	sl := stat.RangeFinite(start, end)
+	return self.SelectRows(sl)
+}
+
+// SelectRows 选择一段记录
 func (self DataFrame) SelectRows(p stat.ScopeLimit) DataFrame {
 	columns := []stat.Series{}
 	for i := range self.columns {
-- 
Gitee


From d16354c389c93c8dbf65ead4d2d782c9bb1d22f9 Mon Sep 17 00:00:00 2001
From: wangfeng <wangfengxy@sina.cn>
Date: Wed, 19 Apr 2023 12:52:39 +0800
Subject: [PATCH 2/3] =?UTF-8?q?=E4=BF=AE=E5=A4=8Dprint=E5=9C=A8=E8=AE=B0?=
 =?UTF-8?q?=E5=BD=95=E9=9B=86=E5=B0=91=E4=BA=8EmaxRows=E5=BA=8F=E5=8F=B7?=
 =?UTF-8?q?=E9=94=991=E4=BD=8D=E7=9A=84bug?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 dataframe_xstring.go | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/dataframe_xstring.go b/dataframe_xstring.go
index 78fbd08..c21015d 100644
--- a/dataframe_xstring.go
+++ b/dataframe_xstring.go
@@ -36,7 +36,7 @@ func (self DataFrame) print(
 		str = fmt.Sprintf("%s error: %v", class, self.Err)
 		return
 	}
-	nMinRows := int(maxRows / 2)
+	nMinRows := maxRows / 2
 	nTotal := 0
 	nrows, ncols := self.Dims()
 	if nrows == 0 || ncols == 0 {
@@ -80,10 +80,10 @@ func (self DataFrame) print(
 		} else if i == nMinRows+1 && shortening {
 			// 跳过
 			rowNumbersOffset -= 1
-		} else if i < nMinRows+1 {
-			add = strconv.Itoa(i-1+rowNumbersOffset) + ":"
-		} else {
+		} else if i >= nMinRows+1 && shortening {
 			add = strconv.Itoa(nrows-maxRows+i-1+rowNumbersOffset) + ":"
+		} else {
+			add = strconv.Itoa(i-1+rowNumbersOffset) + ":"
 		}
 		//fmt.Println(i)
 		records[i] = append([]string{add}, records[i]...)
-- 
Gitee


From db426e1943587b6263cb512c39f66d694d435ce6 Mon Sep 17 00:00:00 2001
From: wangfeng <wangfengxy@sina.cn>
Date: Wed, 19 Apr 2023 12:53:00 +0800
Subject: [PATCH 3/3] =?UTF-8?q?=E5=AE=9E=E7=8E=B0=E5=88=86=E7=BB=84?=
 =?UTF-8?q?=E5=8A=9F=E8=83=BD?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 dataframe_group.go      | 28 ++++++++++++++++++++++++++++
 dataframe_group_test.go | 40 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 68 insertions(+)
 create mode 100644 dataframe_group.go
 create mode 100644 dataframe_group_test.go

diff --git a/dataframe_group.go b/dataframe_group.go
new file mode 100644
index 0000000..e464447
--- /dev/null
+++ b/dataframe_group.go
@@ -0,0 +1,28 @@
+package pandas
+
+import (
+	"gitee.com/quant1x/pandas/stat"
+)
+
+// Group 分组
+func (self DataFrame) Group(columnName string, filter func(kind stat.Type, e any) bool) DataFrame {
+	series := self.Col(columnName)
+	if series.Len() == 0 {
+		return self
+	}
+	t := series.Type()
+	indexes := []int{}
+	series.Apply(func(idx int, v any) {
+		ok := filter(t, v)
+		if ok {
+			indexes = append(indexes, idx)
+		}
+	})
+	ranges := stat.IntsToRanges(indexes)
+	df := DataFrame{}
+	for _, r := range ranges {
+		tmp := self.SelectRows(r)
+		df = df.Concat(tmp)
+	}
+	return df
+}
diff --git a/dataframe_group_test.go b/dataframe_group_test.go
new file mode 100644
index 0000000..0317de4
--- /dev/null
+++ b/dataframe_group_test.go
@@ -0,0 +1,40 @@
+package pandas
+
+import (
+	"fmt"
+	"gitee.com/quant1x/pandas/stat"
+	"testing"
+)
+
+func TestDataFrame_Group(t *testing.T) {
+	type testStruct struct {
+		A string
+		B int
+		C bool
+		D float64
+	}
+	data := []testStruct{
+		{"a", 1, true, 0.0},
+		{"b", 2, false, 0.5},
+		{"b", 3, false, 3.5},
+		{"b", 4, false, 2.5},
+		{"b", 5, false, 1.5},
+		{"a", 6, true, 0.0},
+		{"a", 7, true, 0.0},
+		{"a", 8, true, 0.0},
+		{"a", 9, true, 0.0},
+		{"a", 10, true, 0.0},
+		{"a", 11, true, 0.0},
+	}
+	df := LoadStructs(data)
+	fmt.Println(df)
+
+	df1 := df.Group("A", func(kind stat.Type, e any) bool {
+		v := stat.AnyToString(e)
+		if v == "b" {
+			return true
+		}
+		return false
+	})
+	fmt.Println(df1)
+}
-- 
Gitee