/
dataframe.go
271 lines (217 loc) · 5.66 KB
/
dataframe.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
package dataframe
import (
"sync"
)
type DataFrame struct {
lock sync.RWMutex
Series []Series
n int // Number of rows
}
// NewDataFrame creates a new dataframe.
func NewDataFrame(se ...Series) *DataFrame {
df := &DataFrame{
Series: []Series{},
}
if len(se) > 0 {
var count *int
names := map[string]struct{}{}
for _, s := range se {
if count == nil {
count = &[]int{s.NRows()}[0]
names[s.Name()] = struct{}{}
} else {
if *count != s.NRows() {
panic("different number of rows in series")
}
if _, exists := names[s.Name()]; exists {
panic("names of series must be unique")
}
names[s.Name()] = struct{}{}
}
df.Series = append(df.Series, s)
}
df.n = *count
}
return df
}
func (df *DataFrame) NRows() int {
df.lock.RLock()
defer df.lock.RUnlock()
return df.n
}
// ValuesOptions is used to modify the behaviour of Values()
type ValuesOptions struct {
// InitialRow represents the starting value for iterating
InitialRow int
// Step represents by how much each iteration should step by.
// It can be negative to represent iterating in backwards direction.
// InitialRow should be adjusted to NRows()-1 if Step is negative.
// If Step is 0, the function will panic.
Step int
}
// Values will return an iterator that can be used to iterate through all the values
func (df *DataFrame) Values(options ...ValuesOptions) func() (*int, map[interface{}]interface{}) {
var row int
var step int = 1
if len(options) > 0 {
row = options[0].InitialRow
step = options[0].Step
if step == 0 {
panic("Step can not be zero")
}
}
return func() (*int, map[interface{}]interface{}) {
df.lock.RLock()
defer df.lock.RUnlock()
if row > df.NRows()-1 || row < 0 {
// Don't iterate further
return nil, nil
}
out := map[interface{}]interface{}{}
for idx, aSeries := range df.Series {
val := aSeries.Value(row)
out[aSeries.Name()] = val
out[idx] = val
}
row = row + step
return &[]int{row - step}[0], out
}
}
// Prepend inserts a row at the beginning.
func (df *DataFrame) Prepend(vals ...interface{}) {
df.lock.Lock()
defer df.lock.Unlock()
if len(vals) > 0 {
switch v := vals[0].(type) {
case map[string]interface{}:
names := map[string]struct{}{}
for name, _ := range v {
names[name] = struct{}{}
}
// Check if number of vals is equal to number of series
if len(names) != len(df.Series) {
panic("no. of args not equal to no. of series")
}
for name, val := range v {
df.Series[df.NameToColumn(name)].Prepend(val)
}
default:
// Check if number of vals is equal to number of series
if len(vals) != len(df.Series) {
panic("no. of args not equal to no. of series")
}
for idx, val := range vals {
df.Series[idx].Prepend(val)
}
}
df.n++
}
}
// Append inserts a row at the end.
func (df *DataFrame) Append(vals ...interface{}) {
df.Insert(df.n, vals...)
}
// Insert adds a row to a particular position.
func (df *DataFrame) Insert(row int, vals ...interface{}) {
df.lock.Lock()
defer df.lock.Unlock()
df.insert(row, vals...)
}
func (df *DataFrame) insert(row int, vals ...interface{}) {
if len(vals) > 0 {
switch v := vals[0].(type) {
case map[string]interface{}:
names := map[string]struct{}{}
for name, _ := range v {
names[name] = struct{}{}
}
// Check if number of vals is equal to number of series
if len(names) != len(df.Series) {
panic("no. of args not equal to no. of series")
}
for name, val := range v {
df.Series[df.NameToColumn(name)].Insert(row, val)
}
default:
// Check if number of vals is equal to number of series
if len(vals) != len(df.Series) {
panic("no. of args not equal to no. of series")
}
for idx, val := range vals {
df.Series[idx].Insert(row, val)
}
}
df.n++
}
}
// Remove deletes a row.
func (df *DataFrame) Remove(row int) {
df.lock.Lock()
defer df.lock.Unlock()
for i := range df.Series {
df.Series[i].Remove(row)
}
df.n--
}
// Update is used to update a specific entry.
// col can the name of the series or the column number
func (df *DataFrame) Update(row int, col interface{}, val interface{}) {
df.lock.Lock()
defer df.lock.Unlock()
switch name := col.(type) {
case string:
col = df.NameToColumn(name)
}
df.Series[col.(int)].Update(row, val)
}
// UpdateRow will update an entire row
func (df *DataFrame) UpdateRow(row int, vals ...interface{}) {
df.lock.Lock()
defer df.lock.Unlock()
if len(vals) > 0 {
switch v := vals[0].(type) {
case map[string]interface{}:
for name, val := range v {
df.Series[df.NameToColumn(name)].Update(row, val)
}
default:
// Check if number of vals is equal to number of series
if len(vals) != len(df.Series) {
panic("no. of args not equal to no. of series")
}
for idx, val := range vals {
df.Series[idx].Update(row, val)
}
}
}
}
// NameToColumn returns the index of the series based on the name.
// The starting index is 0.
func (df *DataFrame) NameToColumn(seriesName string) int {
for idx, aSeries := range df.Series {
if aSeries.Name() == seriesName {
return idx
}
}
panic("no series contains name")
}
// Swap is used to swap 2 values based on their row position.
func (df *DataFrame) Swap(row1, row2 int) {
df.lock.Lock()
defer df.lock.Unlock()
df.swap(row1, row2)
}
func (df *DataFrame) swap(row1, row2 int) {
for idx := range df.Series {
df.Series[idx].Swap(row1, row2)
}
}
// Lock will lock the dataframe allowing you to directly manipulate
// the underlying series with confidence.
func (df *DataFrame) Lock() {
df.lock.Lock()
}
// Unlock will unlock the dataframe that was previously locked.
func (df *DataFrame) Unlock() {
df.lock.Unlock()
}