github.com/matrixorigin/matrixone@v0.7.0/pkg/vm/engine/tae/dataio/blockio/reader.go (about)

     1  // Copyright 2021 Matrix Origin
     2  //
     3  // Licensed under the Apache License, Version 2.0 (the "License");
     4  // you may not use this file except in compliance with the License.
     5  // You may obtain a copy of the License at
     6  //
     7  //      http://www.apache.org/licenses/LICENSE-2.0
     8  //
     9  // Unless required by applicable law or agreed to in writing, software
    10  // distributed under the License is distributed on an "AS IS" BASIS,
    11  // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
    12  // See the License for the specific language governing permissions and
    13  // limitations under the License.
    14  
    15  package blockio
    16  
    17  import (
    18  	"context"
    19  	"errors"
    20  	"io"
    21  
    22  	"github.com/google/uuid"
    23  	"github.com/matrixorigin/matrixone/pkg/common/mpool"
    24  	"github.com/matrixorigin/matrixone/pkg/fileservice"
    25  
    26  	"github.com/matrixorigin/matrixone/pkg/container/types"
    27  	"github.com/matrixorigin/matrixone/pkg/container/vector"
    28  	"github.com/matrixorigin/matrixone/pkg/objectio"
    29  	"github.com/matrixorigin/matrixone/pkg/vm/engine/tae/containers"
    30  	"github.com/matrixorigin/matrixone/pkg/vm/engine/tae/tasks"
    31  )
    32  
    33  type Reader struct {
    34  	reader  objectio.Reader
    35  	fs      *objectio.ObjectFS
    36  	key     string
    37  	meta    *Meta
    38  	name    string
    39  	locs    []objectio.Extent
    40  	readCxt context.Context
    41  }
    42  
    43  func NewReader(cxt context.Context, fs *objectio.ObjectFS, key string) (*Reader, error) {
    44  	meta, err := DecodeMetaLocToMeta(key)
    45  	if err != nil {
    46  		return nil, err
    47  	}
    48  	reader, err := objectio.NewObjectReader(meta.GetKey(), fs.Service)
    49  	if err != nil {
    50  		return nil, err
    51  	}
    52  	return &Reader{
    53  		fs:      fs,
    54  		reader:  reader,
    55  		key:     key,
    56  		meta:    meta,
    57  		readCxt: cxt,
    58  	}, nil
    59  }
    60  
    61  func NewCheckpointReader(cxt context.Context, fs fileservice.FileService, key string) (*Reader, error) {
    62  	name, locs, err := DecodeMetaLocToMetas(key)
    63  	if err != nil {
    64  		return nil, err
    65  	}
    66  	reader, err := objectio.NewObjectReader(name, fs)
    67  	if err != nil {
    68  		return nil, err
    69  	}
    70  	return &Reader{
    71  		fs:      objectio.NewObjectFS(fs, ""),
    72  		reader:  reader,
    73  		key:     key,
    74  		name:    name,
    75  		locs:    locs,
    76  		readCxt: cxt,
    77  	}, nil
    78  }
    79  
    80  func (r *Reader) BlkColumnByMetaLoadJob(
    81  	colTypes []types.Type,
    82  	colNames []string,
    83  	nullables []bool,
    84  	block objectio.BlockObject,
    85  ) *tasks.Job {
    86  	exec := func(_ context.Context) (result *tasks.JobResult) {
    87  		bat, err := r.LoadBlkColumnsByMeta(colTypes, colNames, nullables, block)
    88  		return &tasks.JobResult{
    89  			Err: err,
    90  			Res: bat,
    91  		}
    92  	}
    93  	return tasks.NewJob(uuid.NewString(), r.readCxt, exec)
    94  }
    95  
    96  func (r *Reader) BlkColumnsByMetaAndIdxLoadJob(
    97  	colTypes []types.Type,
    98  	colNames []string,
    99  	nullables []bool,
   100  	block objectio.BlockObject,
   101  	idx int,
   102  ) *tasks.Job {
   103  	exec := func(_ context.Context) (result *tasks.JobResult) {
   104  		bat, err := r.LoadBlkColumnsByMetaAndIdx(
   105  			colTypes,
   106  			colNames,
   107  			nullables,
   108  			block,
   109  			idx)
   110  		return &tasks.JobResult{
   111  			Err: err,
   112  			Res: bat,
   113  		}
   114  	}
   115  	return tasks.NewJob(uuid.NewString(), r.readCxt, exec)
   116  }
   117  
   118  func (r *Reader) LoadBlkColumnsByMeta(
   119  	colTypes []types.Type,
   120  	colNames []string,
   121  	nullables []bool,
   122  	block objectio.BlockObject) (*containers.Batch, error) {
   123  	bat := containers.NewBatch()
   124  	if block.GetExtent().End() == 0 {
   125  		return bat, nil
   126  	}
   127  	idxs := make([]uint16, len(colNames))
   128  	for i := range colNames {
   129  		idxs[i] = uint16(i)
   130  	}
   131  	ioResult, err := r.reader.Read(r.readCxt, block.GetExtent(), idxs, nil)
   132  	if err != nil {
   133  		return nil, err
   134  	}
   135  
   136  	for i := range colNames {
   137  		pkgVec := vector.New(colTypes[i])
   138  		data := make([]byte, len(ioResult.Entries[i].Object.([]byte)))
   139  		copy(data, ioResult.Entries[i].Object.([]byte))
   140  		if err = pkgVec.Read(data); err != nil && !errors.Is(err, io.EOF) {
   141  			return bat, err
   142  		}
   143  		var vec containers.Vector
   144  		if pkgVec.Length() == 0 {
   145  			vec = containers.MakeVector(colTypes[i], nullables[i])
   146  		} else {
   147  			vec = containers.NewVectorWithSharedMemory(pkgVec, nullables[i])
   148  		}
   149  		bat.AddVector(colNames[i], vec)
   150  		bat.Vecs[i] = vec
   151  
   152  	}
   153  	return bat, nil
   154  }
   155  
   156  func (r *Reader) LoadBlkColumnsByMetaAndIdx(
   157  	colTypes []types.Type,
   158  	colNames []string,
   159  	nullables []bool,
   160  	block objectio.BlockObject,
   161  	idx int) (*containers.Batch, error) {
   162  	bat := containers.NewBatch()
   163  
   164  	if block.GetExtent().End() == 0 {
   165  		return nil, nil
   166  	}
   167  	col, err := block.GetColumn(uint16(idx))
   168  	if err != nil {
   169  		return bat, err
   170  	}
   171  	data, err := col.GetData(r.readCxt, nil)
   172  	if err != nil {
   173  		return bat, err
   174  	}
   175  	pkgVec := vector.New(colTypes[0])
   176  	v := make([]byte, len(data.Entries[0].Object.([]byte)))
   177  	copy(v, data.Entries[0].Object.([]byte))
   178  	if err = pkgVec.Read(v); err != nil && !errors.Is(err, io.EOF) {
   179  		return bat, err
   180  	}
   181  	var vec containers.Vector
   182  	if pkgVec.Length() == 0 {
   183  		vec = containers.MakeVector(colTypes[0], nullables[0])
   184  	} else {
   185  		vec = containers.NewVectorWithSharedMemory(pkgVec, nullables[0])
   186  	}
   187  	bat.AddVector(colNames[0], vec)
   188  	return bat, nil
   189  }
   190  
   191  func (r *Reader) ReadMeta(m *mpool.MPool) (objectio.BlockObject, error) {
   192  	extents := make([]objectio.Extent, 1)
   193  	extents[0] = r.meta.GetLoc()
   194  	block, err := r.reader.ReadMeta(r.readCxt, extents, m)
   195  	if err != nil {
   196  		return nil, err
   197  	}
   198  	return block[0], err
   199  }
   200  
   201  func (r *Reader) ReadMetas(m *mpool.MPool) ([]objectio.BlockObject, error) {
   202  	block, err := r.reader.ReadMeta(r.readCxt, r.locs, m)
   203  	return block, err
   204  }
   205  
   206  func (r *Reader) GetDataObject(idx uint16, m *mpool.MPool) objectio.ColumnObject {
   207  	block, err := r.ReadMeta(m)
   208  	if err != nil {
   209  		panic(any(err))
   210  	}
   211  	if block == nil {
   212  		return nil
   213  	}
   214  	object, err := block.GetColumn(idx)
   215  	if err != nil {
   216  		panic(any(err))
   217  	}
   218  	return object
   219  }