-
Notifications
You must be signed in to change notification settings - Fork 153
/
rowreader.go
128 lines (105 loc) · 3.54 KB
/
rowreader.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
package bigtable
import (
"fmt"
"cloud.google.com/go/bigtable"
"github.com/influxdata/flux"
"github.com/influxdata/flux/execute"
"github.com/influxdata/flux/semantic"
"github.com/influxdata/flux/values"
"golang.org/x/net/context"
)
type BigtableRowReader struct {
rowsByFamily map[string][]FamilyRow
families []string
familyIndex int
columnIndices map[string]int
rowIndex int
// We need columnNames because maps aren't ordered in Go
columnNames []string
}
func (m *BigtableRowReader) Next() bool {
m.rowIndex++
return m.rowIndex < len(m.rowsByFamily[m.families[m.familyIndex]])
}
func (m *BigtableRowReader) GetNextRow() ([]values.Value, error) {
r := m.rowsByFamily[m.families[m.familyIndex]][m.rowIndex]
if len(r) <= 0 {
return nil, nil
}
// Start off with the row name and time
rowValues := make([]values.Value, len(m.columnIndices))
rowValues[0] = values.NewString(r[0].Row)
rowValues[1] = values.NewTime(values.ConvertTime(r[0].Timestamp.Time()))
rowValues[2] = values.NewString(m.families[m.familyIndex])
// Keeps track of which columnIndices this row includes
encountered := make([]bool, len(m.columnIndices))
// Row key and time have already been appended
encountered[0], encountered[1], encountered[2] = true, true, true
for _, item := range r {
label := item.Column
idx, ok := m.columnIndices[label]
if !ok {
idx = len(m.columnIndices)
m.columnIndices[label] = idx
m.columnNames = append(m.columnNames, label)
encountered = append(encountered, true)
rowValues = append(rowValues, values.NewString(string(item.Value)))
} else {
encountered[idx] = true
rowValues[idx] = values.NewString(string(item.Value))
}
}
for _, idx := range m.columnIndices {
if !encountered[idx] {
rowValues[idx] = values.NewNull(semantic.BasicString)
}
}
return rowValues, nil
}
func (m *BigtableRowReader) ColumnNames() []string {
return m.columnNames
}
func (m *BigtableRowReader) ColumnTypes() []flux.ColType {
return nil
}
func (m *BigtableRowReader) SetColumns([]interface{}) {}
func (m *BigtableRowReader) Close() error { return nil }
func NewBigtableRowReader(ctx context.Context, c *BigtableDecoder) (execute.RowReader, error) {
reader := &BigtableRowReader{
rowsByFamily: make(map[string][]FamilyRow),
families: make([]string, 0),
}
reader.familyIndex = -1
// bigtable.InfiniteRange(start) returns the RowRange consisting of all keys at least as large as start
// "" is the lexicographically smallest byte string - gets all rows
if c.spec.RowSet == nil {
c.spec.RowSet = bigtable.InfiniteRange("")
}
c.spec.ReadOptions = append(c.spec.ReadOptions, bigtable.RowFilter(c.spec.Filter))
if err := c.tbl.ReadRows(ctx, c.spec.RowSet, func(r bigtable.Row) bool {
for family := range r {
if _, ok := reader.rowsByFamily[family]; !ok {
reader.rowsByFamily[family] = make([]FamilyRow, 0)
reader.families = append(reader.families, family)
}
reader.rowsByFamily[family] = append(reader.rowsByFamily[family], r[family])
}
return true
}, c.spec.ReadOptions...); err != nil {
return nil, err
}
return reader, nil
}
func (m *BigtableRowReader) nextFamily() (bool, error) {
m.familyIndex++
m.rowIndex = -1
m.columnIndices = map[string]int{"rowKey": 0, execute.DefaultTimeColLabel: 1, "family": 2}
m.columnNames = []string{"rowKey", execute.DefaultTimeColLabel, "family"}
if len(m.rowsByFamily) == 0 {
return false, fmt.Errorf("no rows found")
}
return m.familyIndex < len(m.rowsByFamily), nil
}
func (m *BigtableRowReader) currentFamily() string {
return m.families[m.familyIndex]
}