feast/go/types/typeconversion.go at java-proto · feast-dev/feast · GitHub

History

346 lines (324 loc) · 11.3 KB

Raw

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

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

package types

import (

"fmt"

"github.com/apache/arrow/go/v17/arrow"

"github.com/apache/arrow/go/v17/arrow/array"

"github.com/apache/arrow/go/v17/arrow/memory"

"github.com/feast-dev/feast/go/protos/feast/types"

)

func ProtoTypeToArrowType(sample *types.Value) (arrow.DataType, error) {

if sample.Val == nil {

return nil, nil

}

switch sample.Val.(type) {

case *types.Value_BytesVal:

return arrow.BinaryTypes.Binary, nil

case *types.Value_StringVal:

return arrow.BinaryTypes.String, nil

case *types.Value_Int32Val:

return arrow.PrimitiveTypes.Int32, nil

case *types.Value_Int64Val:

return arrow.PrimitiveTypes.Int64, nil

case *types.Value_FloatVal:

return arrow.PrimitiveTypes.Float32, nil

case *types.Value_DoubleVal:

return arrow.PrimitiveTypes.Float64, nil

case *types.Value_BoolVal:

return arrow.FixedWidthTypes.Boolean, nil

case *types.Value_BoolListVal:

return arrow.ListOf(arrow.FixedWidthTypes.Boolean), nil

case *types.Value_StringListVal:

return arrow.ListOf(arrow.BinaryTypes.String), nil

case *types.Value_BytesListVal:

return arrow.ListOf(arrow.BinaryTypes.Binary), nil

case *types.Value_Int32ListVal:

return arrow.ListOf(arrow.PrimitiveTypes.Int32), nil

case *types.Value_Int64ListVal:

return arrow.ListOf(arrow.PrimitiveTypes.Int64), nil

case *types.Value_FloatListVal:

return arrow.ListOf(arrow.PrimitiveTypes.Float32), nil

case *types.Value_DoubleListVal:

return arrow.ListOf(arrow.PrimitiveTypes.Float64), nil

case *types.Value_UnixTimestampVal:

return arrow.FixedWidthTypes.Timestamp_s, nil

case *types.Value_UnixTimestampListVal:

return arrow.ListOf(arrow.FixedWidthTypes.Timestamp_s), nil

default:

return nil,

fmt.Errorf("unsupported proto type in proto to arrow conversion: %s", sample.Val)

}

func ValueTypeEnumToArrowType(t types.ValueType_Enum) (arrow.DataType, error) {

switch t {

case types.ValueType_BYTES:

return arrow.BinaryTypes.Binary, nil

case types.ValueType_STRING:

return arrow.BinaryTypes.String, nil

case types.ValueType_INT32:

return arrow.PrimitiveTypes.Int32, nil

case types.ValueType_INT64:

return arrow.PrimitiveTypes.Int64, nil

case types.ValueType_FLOAT:

return arrow.PrimitiveTypes.Float32, nil

case types.ValueType_DOUBLE:

return arrow.PrimitiveTypes.Float64, nil

case types.ValueType_BOOL:

return arrow.FixedWidthTypes.Boolean, nil

case types.ValueType_BOOL_LIST:

return arrow.ListOf(arrow.FixedWidthTypes.Boolean), nil

case types.ValueType_STRING_LIST:

return arrow.ListOf(arrow.BinaryTypes.String), nil

case types.ValueType_BYTES_LIST:

return arrow.ListOf(arrow.BinaryTypes.Binary), nil

case types.ValueType_INT32_LIST:

return arrow.ListOf(arrow.PrimitiveTypes.Int32), nil

case types.ValueType_INT64_LIST:

return arrow.ListOf(arrow.PrimitiveTypes.Int64), nil

case types.ValueType_FLOAT_LIST:

return arrow.ListOf(arrow.PrimitiveTypes.Float32), nil

case types.ValueType_DOUBLE_LIST:

return arrow.ListOf(arrow.PrimitiveTypes.Float64), nil

case types.ValueType_UNIX_TIMESTAMP:

return arrow.FixedWidthTypes.Timestamp_s, nil

case types.ValueType_UNIX_TIMESTAMP_LIST:

return arrow.ListOf(arrow.FixedWidthTypes.Timestamp_s), nil

default:

return nil,

fmt.Errorf("unsupported value type enum in enum to arrow type conversion: %s", t)

}

func CopyProtoValuesToArrowArray(builder array.Builder, values []*types.Value) error {

for _, value := range values {

if value == nil || value.Val == nil {

builder.AppendNull()

continue

}

switch fieldBuilder := builder.(type) {

case *array.BooleanBuilder:

fieldBuilder.Append(value.GetBoolVal())

case *array.BinaryBuilder:

fieldBuilder.Append(value.GetBytesVal())

case *array.StringBuilder:

fieldBuilder.Append(value.GetStringVal())

case *array.Int32Builder:

fieldBuilder.Append(value.GetInt32Val())

case *array.Int64Builder:

fieldBuilder.Append(value.GetInt64Val())

case *array.Float32Builder:

fieldBuilder.Append(value.GetFloatVal())

case *array.Float64Builder:

fieldBuilder.Append(value.GetDoubleVal())

case *array.TimestampBuilder:

fieldBuilder.Append(arrow.Timestamp(value.GetUnixTimestampVal()))

case *array.ListBuilder:

fieldBuilder.Append(true)

switch valueBuilder := fieldBuilder.ValueBuilder().(type) {

case *array.BooleanBuilder:

for _, v := range value.GetBoolListVal().GetVal() {

valueBuilder.Append(v)

}

case *array.BinaryBuilder:

for _, v := range value.GetBytesListVal().GetVal() {

valueBuilder.Append(v)

}

case *array.StringBuilder:

for _, v := range value.GetStringListVal().GetVal() {

valueBuilder.Append(v)

}

case *array.Int32Builder:

for _, v := range value.GetInt32ListVal().GetVal() {

valueBuilder.Append(v)

}

case *array.Int64Builder:

for _, v := range value.GetInt64ListVal().GetVal() {

valueBuilder.Append(v)

}

case *array.Float32Builder:

for _, v := range value.GetFloatListVal().GetVal() {

valueBuilder.Append(v)

}

case *array.Float64Builder:

for _, v := range value.GetDoubleListVal().GetVal() {

valueBuilder.Append(v)

}

case *array.TimestampBuilder:

for _, v := range value.GetUnixTimestampListVal().GetVal() {

valueBuilder.Append(arrow.Timestamp(v))

}

default:

return fmt.Errorf("unsupported array builder: %s", builder)

}

return nil

}

func ArrowValuesToProtoValues(arr arrow.Array) ([]*types.Value, error) {

values := make([]*types.Value, 0)

if listArr, ok := arr.(*array.List); ok {

listValues := listArr.ListValues()

offsets := listArr.Offsets()[1:]

pos := 0

for idx := 0; idx < listArr.Len(); idx++ {

switch listValues.DataType() {

case arrow.PrimitiveTypes.Int32:

vals := make([]int32, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = listValues.(*array.Int32).Value(j)

}

values = append(values,

&types.Value{Val: &types.Value_Int32ListVal{Int32ListVal: &types.Int32List{Val: vals}}})

case arrow.PrimitiveTypes.Int64:

vals := make([]int64, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = listValues.(*array.Int64).Value(j)

}

values = append(values,

&types.Value{Val: &types.Value_Int64ListVal{Int64ListVal: &types.Int64List{Val: vals}}})

case arrow.PrimitiveTypes.Float32:

vals := make([]float32, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = listValues.(*array.Float32).Value(j)

}

values = append(values,

&types.Value{Val: &types.Value_FloatListVal{FloatListVal: &types.FloatList{Val: vals}}})

case arrow.PrimitiveTypes.Float64:

vals := make([]float64, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = listValues.(*array.Float64).Value(j)

}

values = append(values,

&types.Value{Val: &types.Value_DoubleListVal{DoubleListVal: &types.DoubleList{Val: vals}}})

case arrow.BinaryTypes.Binary:

vals := make([][]byte, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = listValues.(*array.Binary).Value(j)

}

values = append(values,

&types.Value{Val: &types.Value_BytesListVal{BytesListVal: &types.BytesList{Val: vals}}})

case arrow.BinaryTypes.String:

vals := make([]string, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = listValues.(*array.String).Value(j)

}

values = append(values,

&types.Value{Val: &types.Value_StringListVal{StringListVal: &types.StringList{Val: vals}}})

case arrow.FixedWidthTypes.Boolean:

vals := make([]bool, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = listValues.(*array.Boolean).Value(j)

}

values = append(values,

&types.Value{Val: &types.Value_BoolListVal{BoolListVal: &types.BoolList{Val: vals}}})

case arrow.FixedWidthTypes.Timestamp_s:

vals := make([]int64, int(offsets[idx])-pos)

for j := pos; j < int(offsets[idx]); j++ {

vals[j-pos] = int64(listValues.(*array.Timestamp).Value(j))

}

values = append(values,

&types.Value{Val: &types.Value_UnixTimestampListVal{

UnixTimestampListVal: &types.Int64List{Val: vals}}})

}

// set the end of current element as start of the next

pos = int(offsets[idx])

}

return values, nil

}

switch arr.DataType() {

case arrow.PrimitiveTypes.Int32:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_Int32Val{Int32Val: arr.(*array.Int32).Value(idx)}})

}

case arrow.PrimitiveTypes.Int64:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_Int64Val{Int64Val: arr.(*array.Int64).Value(idx)}})

}

case arrow.PrimitiveTypes.Float32:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_FloatVal{FloatVal: arr.(*array.Float32).Value(idx)}})

}

case arrow.PrimitiveTypes.Float64:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_DoubleVal{DoubleVal: arr.(*array.Float64).Value(idx)}})

}

case arrow.FixedWidthTypes.Boolean:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_BoolVal{BoolVal: arr.(*array.Boolean).Value(idx)}})

}

case arrow.BinaryTypes.Binary:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_BytesVal{BytesVal: arr.(*array.Binary).Value(idx)}})

}

case arrow.BinaryTypes.String:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_StringVal{StringVal: arr.(*array.String).Value(idx)}})

}

case arrow.FixedWidthTypes.Timestamp_s:

for idx := 0; idx < arr.Len(); idx++ {

if arr.IsNull(idx) {

values = append(values, &types.Value{})

} else {

values = append(values, &types.Value{Val: &types.Value_UnixTimestampVal{UnixTimestampVal: int64(arr.(*array.Timestamp).Value(idx))}})

}

case arrow.Null:

for idx := 0; idx < arr.Len(); idx++ {

values = append(values, &types.Value{})

}

default:

return nil, fmt.Errorf("unsupported arrow to proto conversion for type %s", arr.DataType())

}

return values, nil

}

func ProtoValuesToArrowArray(protoValues []*types.Value, arrowAllocator memory.Allocator, numRows int) (arrow.Array, error) {

var fieldType arrow.DataType

var err error

for _, val := range protoValues {

if val != nil {

fieldType, err = ProtoTypeToArrowType(val)

if err != nil {

return nil, err

}

if fieldType != nil {

break

}

if fieldType != nil {

builder := array.NewBuilder(arrowAllocator, fieldType)

err = CopyProtoValuesToArrowArray(builder, protoValues)

if err != nil {

return nil, err

}

return builder.NewArray(), nil

} else {

return array.NewNull(numRows), nil

}

feast/go/types/typeconversion.go at java-proto · feast-dev/feast · GitHub

Uh oh!

Expand file tree

Breadcrumbs

typeconversion.go

Latest commit

History

Breadcrumbs

typeconversion.go

File metadata and controls