Files
e2787d3981 enhance: standardize error handling on merr + Sys/Input classification (#50221)
issue: #47420

## What this PR does

Project-wide migration of raw `fmt.Errorf` / `errors.New` in function
bodies onto
the `merr` framework, plus the Sys-vs-Input error classification and the
machinery it drives (retriability, fine-grained metrics, segcore
unification),
plus the convention docs and a linter that keeps it from regressing.

Scope: storage, proxy, coordinators (root/data/query), query node, data
node,
`pkg/util` & `internal/util`, expression parser, message queue,
streaming, and
misc packages. Bare raw-error usages went from ~3000 to a ~340 allowlist
(package-level sentinels / build-tag / test sites).

---

## How to review this PR

It is large but the vast majority is mechanical. Changes fall into three
tiers;
spend review budget on Part 2 and Part 3.

### Part 1 — Mechanical standardization (low risk, verify by rule)

Each converted call follows one of a small fixed set of rules. To
review, check
that each site obeys the matching rule rather than reading every line:

| Pattern | Rule |
|---|---|
| `fmt.Errorf("...")` originating a new error | →
`merr.WrapErrXxxMsg("...")` with a code matching the failure's meaning |
| Adding context to an existing typed error | → `merr.Wrap(err, "...")`
/ `merr.Wrapf(...)` — **preserves** the inner code (never `WrapErr*Err`,
which overwrites it) |
| Errors inside the streaming subsystem | → `status.New*` factories
(StreamingError), **not** merr — this is the component-internal dialect
(see `docs/dev/error_handling_guide.md`) |
| Low-level / control-flow signal caught by `errors.Is` | → kept as a
package-level `errors.New` sentinel (lowercase, same-package) |

Conventions are documented in `docs/dev/error_handling_guide.md`
(how-to) and
`docs/dev/error_sentinel_convention.md` (rules + audit). A
`gocritic`/`ruleguard`
rule (`rawmerrerror`, in `rules.go`) enforces "no raw `return
errors.New/fmt.Errorf`"
under `make verifiers`.

### Part 2 — Behavior changes (review these closely)

These are the sites where the wire contract or runtime behavior changes,
not just
the source text. Listed by category; representative locations given,
full set in
the diff.

**A. gRPC wire-code shifts: `UnexpectedError(1)/Code 65535` → typed
code.**
Where a handler previously returned a raw error (collapsed to
`Code=65535` on the
wire), it now returns a typed merr, so the client sees a real code. The
most
common shift is to `IllegalArgument(5)/Code 1100` (ParameterInvalid).
Touch
points include datanode task handlers (CreateTask/Query/Drop), proxy
Upsert,
querynode GetMetrics, datacoord CreateIndex, httpserver query-response
builder,
and typeutil schema validation. One code refinement: an index-param
validation
moved `1100` → `1101` (ParameterMissing). **Client/SDK assertions and
any code
that switched on `Code=65535` for these paths must be re-checked** (the
go_client
e2e assertions were already aligned in this PR).

**B. Prometheus `status` label contract change (externally visible).**
The proxy metric's coarse `fail` / `rejected` values are split into
`fail_input` / `fail_system` and `rejected_user` / `rejected_system` (in
`requestutil.ParseMetricLabel`; auth/privilege rejections count as
`rejected_user`), so dashboards can attribute a failure to caller vs
operator.
**Dashboards/alerts querying `status="fail"` must migrate to
`status=~"fail_.*"`, and `status="rejected"` to
`status=~"rejected_.*"`.** The
in-repo Grafana dashboard is already migrated; external dashboards built
on the
old values silently go empty after upgrade. This is the one change that
requires an ops-side migration.

**C. Retriability semantics.**
- C1: `merr.Status(err)` now forces `Retriable=false` when the error is
an
`InputError` — a malformed request can never succeed on blind retry, so
clients
never get the self-contradictory "your input is wrong but you may
retry".
- C2: `retry.Do` short-circuits an `InputError` (non-retriable) — **but
only when
  the caller did not pass a `RetryErr` predicate**. The check is an
`if c.isRetryErr != nil { ... } else if InputError { ... }` *mutually
exclusive*
branch (`pkg/util/retry/retry.go`): an explicit `RetryErr` takes
precedence and
bypasses the InputError abort. `retry.Handle` deliberately does **not**
apply
the InputError abort (its callers signal abort via `shouldRetry=false`).
Four
flusher startup callsites that must retry through transient "not ready"
errors
  were given explicit `RetryErr` escape hatches.

**D. segcore (C++→Go) error classification.**
A single shared Go-side table (`pkg/util/merr/segcore.go`) maps each
segcore code
to a merr sentinel + InputError/signal category, replacing scattered
hand-written
`if errorCode == ...` switches in the cgo wrappers. **Wire `Code` values
change
for every segcore pass-through error, not just the remapped ones.**
Named
sentinels remap (C++ `2003` → merr `2001`, `2033` → `2002`,
Folly/Knowhere codes
likewise); **all remaining pass-through codes (`2004`–`2043`, previously
surfaced to clients as raw C++ enum values) now serialize as `2000`**
(`ErrSegcore`), with the original C++ code preserved in the `Reason`
text
(`segcoreCode=...`); unknown/future codes collapse to `2000` as well
(pinned by
the `wire_code_projection` test). Transient segcore classes (object
storage /
file IO / OOM / mmap / FieldNotLoaded — 11 codes) now report
`Retriable=true`.
**Any client switching on raw segcore codes in the `2004`–`2043` range
must be
re-checked**; the in-Reason code remains available for diagnostics.
Signal
codes (PretendFinished / FollyCancel) are recognized centrally.
`errors.Is`-based
control flow on these (e.g. scheduler skip/retry) is preserved.

**E. InputError classification (25 sentinels + dynamic marks).**
25 sentinels in `errors.go` carry `WithErrorType(InputError)` (the
Collection /
ResourceGroup / Database families, `ErrIndexDuplicate`,
`ErrParameterInvalid`,
`ErrPrivilegeNotAuthenticated`, `ErrImportFailed`, `ErrQueryPlan`, ...),
plus dynamic
marks for the 8 segcore input codes (ExprInvalid, DimNotMatch,
MetricTypeInvalid, FieldIDInvalid, ...) and
`WrapErrAsInputError`. The widest blast radius is `ErrParameterInvalid`
(1100):
~2335 `WrapErrParameterInvalid*` callsites now classify as input /
non-retriable. Because of C1/C2 this changes retriability for
any path that returns these. **The audit to confirm no transient path
was
mis-marked is the single most important review item** (see Part 3). One
reverse
correction: storage field-stats parsing moved from `ErrParameterInvalid`
(input)
to `ErrDataIntegrity` — a corrupted stored stat is data corruption, not
user
input.

### Part 3 — Known risks & traps (called out proactively)

1. **`merr.Wrap` vs `WrapErr*Err` (code-masking).** `WrapErr*Err` builds
a
`wrappedMilvusError{sentinel: ErrServiceInternal}` whose `code()`
returns the
*outer* sentinel — it overwrites the inner typed code and hides the
`errors.Is`
chain. This is intentional (use it to *deliberately* downgrade), but it
was a
recurring conversion defect; the rule "add context with `merr.Wrap`,
downgrade
with `WrapErr*Err`" is enforced by convention and reviewed across the
diff.
2. **InputError × `retry.Do` blast radius.** Marking a sentinel
`InputError` makes
any `retry.Do(...)` without a `RetryErr` predicate stop retrying it.
Reviewers
should sanity-check that no transient use of the 19 newly-marked
sentinels
(especially `ErrParameterInvalid`) sits inside a retry loop that needed
to keep
   spinning. The known flusher cases were handled (see C2).
3. **The ~340 raw-error allowlist.** What remains as bare `errors.New`
is, by
design: package-level sentinels (caught by `errors.Is`), `//go:build
test`
sites, and out-of-band trees (`cmd/`, `tests/`, codegen, walimpls). The
linter
only bans the *direct-return* form; assignment-then-return escapes and
the full
no-exceptions ban are deferred to an AST-based linter (Tier 2,
documented).
4. **segcore C++ second step deferred.** This PR unifies classification
on the Go
side; splitting the dual-semantic C++ codes at the source is a
follow-up.

---

## Validation

- `make verifiers`: Go side clean (gofmt + static-check across modules,
including
  the new `rawmerrerror` rule with a 0-hit baseline repo-wide).
- `make test-go`: passing; the one real regression introduced (a
datanode
`invalid_task_type` assertion shifting `1` → `5` from a ParameterInvalid
  conversion) was fixed in-tree.
- go_client e2e CreateIndex assertions aligned to the new merr messages.

---------

Signed-off-by: zhenshan.cao <zhenshan.cao@zilliz.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-06-12 15:04:51 -07:00

519 lines
17 KiB
Go

// Licensed to the LF AI & Data foundation under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package httpserver
import (
"google.golang.org/protobuf/proto"
"github.com/milvus-io/milvus-proto/go-api/v3/commonpb"
"github.com/milvus-io/milvus-proto/go-api/v3/milvuspb"
"github.com/milvus-io/milvus-proto/go-api/v3/schemapb"
"github.com/milvus-io/milvus/internal/json"
"github.com/milvus-io/milvus/pkg/v3/util/merr"
"github.com/milvus-io/milvus/pkg/v3/util/typeutil"
)
// We wrap original protobuf structure for 2 reasons:
// 1. Milvus uses `bytes` as the type of `schema` field,
// while the bytes has to be serialized by proto.Marshal.
// It's very inconvenient for an HTTP clien to do this,
// so we change the type to a struct,
// and does the conversion for user.
// 2. Some fields uses proto.oneof, does not supported directly json marshal
// so we have to implements the marshal procedure. example: InsertReqeust
// WrappedCreateCollectionRequest wraps CreateCollectionRequest
type WrappedCreateCollectionRequest struct {
// Not useful for now
Base *commonpb.MsgBase `protobuf:"bytes,1,opt,name=base,proto3" json:"base,omitempty"`
// Not useful for now
DbName string `protobuf:"bytes,2,opt,name=db_name,json=dbName,proto3" json:"db_name,omitempty"`
// The unique collection name in milvus.(Required)
CollectionName string `protobuf:"bytes,3,opt,name=collection_name,json=collectionName,proto3" json:"collection_name,omitempty"`
// The serialized `schema.CollectionSchema`(Required)
Schema schemapb.CollectionSchema `protobuf:"bytes,4,opt,name=schema,proto3" json:"schema,omitempty"`
// Once set, no modification is allowed (Optional)
// https://github.com/milvus-io/milvus/issues/6690
ShardsNum int32 `protobuf:"varint,5,opt,name=shards_num,json=shardsNum,proto3" json:"shards_num,omitempty"`
// The consistency level that the collection used, modification is not supported now.
ConsistencyLevel commonpb.ConsistencyLevel `protobuf:"varint,6,opt,name=consistency_level,json=consistencyLevel,proto3,enum=milvus.proto.common.ConsistencyLevel" json:"consistency_level,omitempty"`
Properties []*commonpb.KeyValuePair `protobuf:"bytes,13,rep,name=properties,proto3" json:"properties,omitempty"`
}
// WrappedInsertRequest is the InsertRequest wrapped for RESTful request
type WrappedInsertRequest struct {
Base *commonpb.MsgBase `json:"base,omitempty"`
DbName string `json:"db_name,omitempty"`
CollectionName string `json:"collection_name,omitempty"`
PartitionName string `json:"partition_name,omitempty"`
FieldsData []*FieldData `json:"fields_data,omitempty"`
HashKeys []uint32 `json:"hash_keys,omitempty"`
NumRows uint32 `json:"num_rows,omitempty"`
}
func (w *WrappedInsertRequest) AsInsertRequest() (*milvuspb.InsertRequest, error) {
fieldData, err := convertFieldDataArray(w.FieldsData)
if err != nil {
return nil, badRequestf(err, "convert field data failed")
}
return &milvuspb.InsertRequest{
Base: w.Base,
DbName: w.DbName,
CollectionName: w.CollectionName,
PartitionName: w.PartitionName,
FieldsData: fieldData,
HashKeys: w.HashKeys,
NumRows: w.NumRows,
}, nil
}
// FieldData is the field data in RESTful request that can be convertd to schemapb.FieldData
type FieldData struct {
Type schemapb.DataType `json:"type,omitempty"`
FieldName string `json:"field_name,omitempty"`
Field json.RawMessage `json:"field,omitempty"` // we use postpone the unmarshal until we know the type
FieldID int64 `json:"field_id,omitempty"`
}
func (f *FieldData) makePbFloat16OrBfloat16Array(raw json.RawMessage, serializeFunc func([]float32) []byte) ([]byte, int64, error) {
wrappedData := [][]float32{}
err := json.Unmarshal(raw, &wrappedData)
if err != nil {
return nil, 0, newFieldDataError(f.FieldName, err)
}
if len(wrappedData) < 1 {
return nil, 0, merr.WrapErrParameterInvalidMsg("at least one row for insert")
}
array0 := wrappedData[0]
dim := len(array0)
if dim < 1 {
return nil, 0, merr.WrapErrParameterInvalidMsg("dim must >= 1")
}
data := make([]byte, 0, len(wrappedData)*dim*2)
for _, fp32Array := range wrappedData {
data = append(data, serializeFunc(fp32Array)...)
}
return data, int64(dim), nil
}
// AsSchemapb converts the FieldData to schemapb.FieldData
func (f *FieldData) AsSchemapb() (*schemapb.FieldData, error) {
// is scarlar
ret := schemapb.FieldData{
Type: f.Type,
FieldName: f.FieldName,
FieldId: f.FieldID,
}
raw := f.Field
switch f.Type {
case schemapb.DataType_Bool:
data := []bool{}
err := json.Unmarshal(raw, &data)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
ret.Field = &schemapb.FieldData_Scalars{
Scalars: &schemapb.ScalarField{
Data: &schemapb.ScalarField_BoolData{
BoolData: &schemapb.BoolArray{
Data: data,
},
},
},
}
case schemapb.DataType_VarChar:
data := []string{}
err := json.Unmarshal(raw, &data)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
ret.Field = &schemapb.FieldData_Scalars{
Scalars: &schemapb.ScalarField{
Data: &schemapb.ScalarField_StringData{
StringData: &schemapb.StringArray{
Data: data,
},
},
},
}
case schemapb.DataType_Int8, schemapb.DataType_Int16, schemapb.DataType_Int32:
data := []int32{}
err := json.Unmarshal(raw, &data)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
ret.Field = &schemapb.FieldData_Scalars{
Scalars: &schemapb.ScalarField{
Data: &schemapb.ScalarField_IntData{
IntData: &schemapb.IntArray{
Data: data,
},
},
},
}
case schemapb.DataType_Int64:
data := []int64{}
err := json.Unmarshal(raw, &data)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
ret.Field = &schemapb.FieldData_Scalars{
Scalars: &schemapb.ScalarField{
Data: &schemapb.ScalarField_LongData{
LongData: &schemapb.LongArray{
Data: data,
},
},
},
}
case schemapb.DataType_Float:
data := []float32{}
err := json.Unmarshal(raw, &data)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
ret.Field = &schemapb.FieldData_Scalars{
Scalars: &schemapb.ScalarField{
Data: &schemapb.ScalarField_FloatData{
FloatData: &schemapb.FloatArray{
Data: data,
},
},
},
}
case schemapb.DataType_Double:
data := []float64{}
err := json.Unmarshal(raw, &data)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
ret.Field = &schemapb.FieldData_Scalars{
Scalars: &schemapb.ScalarField{
Data: &schemapb.ScalarField_DoubleData{
DoubleData: &schemapb.DoubleArray{
Data: data,
},
},
},
}
case schemapb.DataType_Timestamptz:
data := []int64{}
err := json.Unmarshal(raw, &data)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
ret.Field = &schemapb.FieldData_Scalars{
Scalars: &schemapb.ScalarField{
Data: &schemapb.ScalarField_TimestamptzData{
TimestamptzData: &schemapb.TimestamptzArray{
Data: data,
},
},
},
}
case schemapb.DataType_FloatVector:
wrappedData := [][]float32{}
err := json.Unmarshal(raw, &wrappedData)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
if len(wrappedData) < 1 {
return nil, merr.WrapErrParameterInvalidMsg("at least one row for insert")
}
array0 := wrappedData[0]
dim := len(array0)
if dim < 1 {
return nil, merr.WrapErrParameterInvalidMsg("dim must >= 1")
}
data := make([]float32, len(wrappedData)*dim)
var i int
for _, dataArray := range wrappedData {
for _, v := range dataArray {
data[i] = v
i++
}
}
ret.Field = &schemapb.FieldData_Vectors{
Vectors: &schemapb.VectorField{
Dim: int64(dim),
Data: &schemapb.VectorField_FloatVector{
FloatVector: &schemapb.FloatArray{
Data: data,
},
},
},
}
case schemapb.DataType_Float16Vector:
// only support float32 conversion right now
data, dim, err := f.makePbFloat16OrBfloat16Array(raw, typeutil.Float32ArrayToFloat16Bytes)
if err != nil {
return nil, err
}
ret.Field = &schemapb.FieldData_Vectors{
Vectors: &schemapb.VectorField{
Dim: dim,
Data: &schemapb.VectorField_Float16Vector{
Float16Vector: data,
},
},
}
case schemapb.DataType_BFloat16Vector:
// only support float32 conversion right now
data, dim, err := f.makePbFloat16OrBfloat16Array(raw, typeutil.Float32ArrayToBFloat16Bytes)
if err != nil {
return nil, err
}
ret.Field = &schemapb.FieldData_Vectors{
Vectors: &schemapb.VectorField{
Dim: dim,
Data: &schemapb.VectorField_Bfloat16Vector{
Bfloat16Vector: data,
},
},
}
case schemapb.DataType_SparseFloatVector:
var wrappedData []map[string]interface{}
err := json.Unmarshal(raw, &wrappedData)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
if len(wrappedData) < 1 {
return nil, merr.WrapErrParameterInvalidMsg("at least one row for insert")
}
data := make([][]byte, len(wrappedData))
dim := int64(0)
for _, row := range wrappedData {
rowData, err := typeutil.CreateSparseFloatRowFromMap(row)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
data = append(data, rowData)
rowDim := typeutil.SparseFloatRowDim(rowData)
if rowDim > dim {
dim = rowDim
}
}
ret.Field = &schemapb.FieldData_Vectors{
Vectors: &schemapb.VectorField{
Dim: dim,
Data: &schemapb.VectorField_SparseFloatVector{
SparseFloatVector: &schemapb.SparseFloatArray{
Dim: dim,
Contents: data,
},
},
},
}
case schemapb.DataType_Int8Vector:
wrappedData := [][]int8{}
err := json.Unmarshal(raw, &wrappedData)
if err != nil {
return nil, newFieldDataError(f.FieldName, err)
}
if len(wrappedData) < 1 {
return nil, merr.WrapErrParameterInvalidMsg("at least one row for insert")
}
array0 := wrappedData[0]
dim := len(array0)
if dim < 1 {
return nil, merr.WrapErrParameterInvalidMsg("dim must >= 1")
}
data := make([]byte, len(wrappedData)*dim)
var i int
for _, dataArray := range wrappedData {
for _, v := range dataArray {
data[i] = byte(v)
i++
}
}
ret.Field = &schemapb.FieldData_Vectors{
Vectors: &schemapb.VectorField{
Dim: int64(dim),
Data: &schemapb.VectorField_Int8Vector{
Int8Vector: data,
},
},
}
default:
return nil, merr.WrapErrParameterInvalidMsg("unsupported data type")
}
return &ret, nil
}
func newFieldDataError(field string, err error) error {
return merr.WrapErrParameterInvalidErr(err, "parse field[%s]", field)
}
func convertFieldDataArray(input []*FieldData) ([]*schemapb.FieldData, error) {
ret := make([]*schemapb.FieldData, len(input))
for i, v := range input {
fieldData, err := v.AsSchemapb()
if err != nil {
return nil, err
}
ret[i] = fieldData
}
return ret, nil
}
// SearchRequest is the RESTful request body for search
type SearchRequest struct {
Base *commonpb.MsgBase `protobuf:"bytes,1,opt,name=base,proto3" json:"base,omitempty"`
DbName string `protobuf:"bytes,2,opt,name=db_name,json=dbName,proto3" json:"db_name,omitempty"`
CollectionName string `protobuf:"bytes,3,opt,name=collection_name,json=collectionName,proto3" json:"collection_name,omitempty"`
PartitionNames []string `protobuf:"bytes,4,rep,name=partition_names,json=partitionNames,proto3" json:"partition_names,omitempty"`
Dsl string `protobuf:"bytes,5,opt,name=dsl,proto3" json:"dsl,omitempty"`
DslType commonpb.DslType `protobuf:"varint,7,opt,name=dsl_type,json=dslType,proto3,enum=milvus.proto.common.DslType" json:"dsl_type,omitempty"`
BinaryVectors [][]byte `json:"binary_vectors,omitempty"`
Vectors [][]float32 `json:"vectors,omitempty"`
OutputFields []string `protobuf:"bytes,8,rep,name=output_fields,json=outputFields,proto3" json:"output_fields,omitempty"`
SearchParams []*commonpb.KeyValuePair `protobuf:"bytes,9,rep,name=search_params,json=searchParams,proto3" json:"search_params,omitempty"`
TravelTimestamp uint64 `protobuf:"varint,10,opt,name=travel_timestamp,json=travelTimestamp,proto3" json:"travel_timestamp,omitempty"`
GuaranteeTimestamp uint64 `protobuf:"varint,11,opt,name=guarantee_timestamp,json=guaranteeTimestamp,proto3" json:"guarantee_timestamp,omitempty"`
Nq int64 `protobuf:"varint,12,opt,name=nq,proto3" json:"nq,omitempty"`
SearchAggregation *SearchAggregationReq `json:"searchAggregation,omitempty"`
SearchAggregationSnake *SearchAggregationReq `json:"search_aggregation,omitempty"`
}
func (r *SearchRequest) HasSearchAggregation() bool {
return r.SearchAggregation != nil || r.SearchAggregationSnake != nil
}
func binaryVector2Bytes(vectors [][]byte) []byte {
ph := &commonpb.PlaceholderValue{
Tag: "$0",
Type: commonpb.PlaceholderType_BinaryVector,
Values: make([][]byte, 0, len(vectors)),
}
ph.Values = append(ph.Values, vectors...)
phg := &commonpb.PlaceholderGroup{
Placeholders: []*commonpb.PlaceholderValue{
ph,
},
}
ret, _ := proto.Marshal(phg)
return ret
}
func vector2Bytes(vectors [][]float32) []byte {
ph := &commonpb.PlaceholderValue{
Tag: "$0",
Type: commonpb.PlaceholderType_FloatVector,
Values: make([][]byte, 0, len(vectors)),
}
for _, vector := range vectors {
ph.Values = append(ph.Values, typeutil.Float32ArrayToBytes(vector))
}
phg := &commonpb.PlaceholderGroup{
Placeholders: []*commonpb.PlaceholderValue{
ph,
},
}
ret, _ := proto.Marshal(phg)
return ret
}
// WrappedCalcDistanceRequest is the RESTful request body for calc distance
type WrappedCalcDistanceRequest struct {
Base *commonpb.MsgBase `protobuf:"bytes,1,opt,name=base,proto3" json:"base,omitempty"`
OpLeft VectorsArray `json:"op_left,omitempty"`
OpRight VectorsArray `json:"op_right,omitempty"`
Params []*commonpb.KeyValuePair `json:"params,omitempty"`
}
// VectorsArray is vector array, assigned by vectors or ids
type VectorsArray struct {
// Dim of vectors or binary_vectors, not needed when use ids
Dim int64 `json:"dim,omitempty"`
// Vectors is an array of vector divided by given dim. Disabled when ids or binary_vectors is set
Vectors []float32 `json:"vectors,omitempty"`
// Vectors is an array of binary vector divided by given dim. Disabled when IDs is set
BinaryVectors []byte `json:"binary_vectors,omitempty"`
// IDs of vector field in milvus, if not nil, vectors will be ignored
IDs *VectorIDs `json:"ids,omitempty"`
}
func (v *VectorsArray) isIDs() bool {
return v.IDs != nil
}
func (v *VectorsArray) isBinaryVector() bool {
return v.IDs == nil && len(v.BinaryVectors) > 0
}
// AsPbVectorArray convert as milvuspb.VectorArray
func (v *VectorsArray) AsPbVectorArray() *milvuspb.VectorsArray {
ret := &milvuspb.VectorsArray{}
switch {
case v.isIDs():
ids := &milvuspb.VectorsArray_IdArray{}
ids.IdArray = &milvuspb.VectorIDs{
CollectionName: v.IDs.CollectionName,
FieldName: v.IDs.FieldName,
}
ids.IdArray.PartitionNames = v.IDs.PartitionNames
ids.IdArray.IdArray = &schemapb.IDs{}
ids.IdArray.IdArray.IdField = &schemapb.IDs_IntId{
IntId: &schemapb.LongArray{
Data: v.IDs.IDArray,
},
}
ret.Array = ids
case v.isBinaryVector():
vf := &schemapb.VectorField{
Dim: v.Dim,
}
vf.Data = &schemapb.VectorField_BinaryVector{
BinaryVector: v.BinaryVectors,
}
ret.Array = &milvuspb.VectorsArray_DataArray{
DataArray: vf,
}
default:
// take it as ordinary vectors
vf := &schemapb.VectorField{
Dim: v.Dim,
}
vf.Data = &schemapb.VectorField_FloatVector{
FloatVector: &schemapb.FloatArray{
Data: v.Vectors,
},
}
ret.Array = &milvuspb.VectorsArray_DataArray{
DataArray: vf,
}
}
return ret
}
// VectorIDs is an array of id reference in milvus
type VectorIDs struct {
CollectionName string `protobuf:"bytes,1,opt,name=collection_name,json=collectionName,proto3" json:"collection_name,omitempty"`
FieldName string `protobuf:"bytes,2,opt,name=field_name,json=fieldName,proto3" json:"field_name,omitempty"`
PartitionNames []string `json:"partition_names"`
IDArray []int64 `json:"id_array,omitempty"`
}