Refactor indexer (#25174)
Refactor `modules/indexer` to make it more maintainable. And it can be
easier to support more features. I'm trying to solve some of issue
searching, this is a precursor to making functional changes.
Current supported engines and the index versions:
| engines | issues | code |
| - | - | - |
| db | Just a wrapper for database queries, doesn't need version | - |
| bleve | The version of index is **2** | The version of index is **6**
|
| elasticsearch | The old index has no version, will be treated as
version **0** in this PR | The version of index is **1** |
| meilisearch | The old index has no version, will be treated as version
**0** in this PR | - |
## Changes
### Split
Splited it into mutiple packages
```text
indexer
├── internal
│ ├── bleve
│ ├── db
│ ├── elasticsearch
│ └── meilisearch
├── code
│ ├── bleve
│ ├── elasticsearch
│ └── internal
└── issues
├── bleve
├── db
├── elasticsearch
├── internal
└── meilisearch
```
- `indexer/interanal`: Internal shared package for indexer.
- `indexer/interanal/[engine]`: Internal shared package for each engine
(bleve/db/elasticsearch/meilisearch).
- `indexer/code`: Implementations for code indexer.
- `indexer/code/internal`: Internal shared package for code indexer.
- `indexer/code/[engine]`: Implementation via each engine for code
indexer.
- `indexer/issues`: Implementations for issues indexer.
### Deduplication
- Combine `Init/Ping/Close` for code indexer and issues indexer.
- ~Combine `issues.indexerHolder` and `code.wrappedIndexer` to
`internal.IndexHolder`.~ Remove it, use dummy indexer instead when the
indexer is not ready.
- Duplicate two copies of creating ES clients.
- Duplicate two copies of `indexerID()`.
### Enhancement
- [x] Support index version for elasticsearch issues indexer, the old
index without version will be treated as version 0.
- [x] Fix spell of `elastic_search/ElasticSearch`, it should be
`Elasticsearch`.
- [x] Improve versioning of ES index. We don't need `Aliases`:
- Gitea does't need aliases for "Zero Downtime" because it never delete
old indexes.
- The old code of issues indexer uses the orignal name to create issue
index, so it's tricky to convert it to an alias.
- [x] Support index version for meilisearch issues indexer, the old
index without version will be treated as version 0.
- [x] Do "ping" only when `Ping` has been called, don't ping
periodically and cache the status.
- [x] Support the context parameter whenever possible.
- [x] Fix outdated example config.
- [x] Give up the requeue logic of issues indexer: When indexing fails,
call Ping to check if it was caused by the engine being unavailable, and
only requeue the task if the engine is unavailable.
- It is fragile and tricky, could cause data losing (It did happen when
I was doing some tests for this PR). And it works for ES only.
- Just always requeue the failed task, if it caused by bad data, it's a
bug of Gitea which should be fixed.
---------
Co-authored-by: Giteabot <teabot@gitea.io>
2023-06-23 08:37:56 -04:00
|
|
|
// Copyright 2019 The Gitea Authors. All rights reserved.
|
|
|
|
// SPDX-License-Identifier: MIT
|
|
|
|
|
|
|
|
package elasticsearch
|
|
|
|
|
|
|
|
import (
|
|
|
|
"context"
|
|
|
|
"fmt"
|
|
|
|
"strconv"
|
|
|
|
|
|
|
|
"code.gitea.io/gitea/modules/graceful"
|
|
|
|
indexer_internal "code.gitea.io/gitea/modules/indexer/internal"
|
|
|
|
inner_elasticsearch "code.gitea.io/gitea/modules/indexer/internal/elasticsearch"
|
|
|
|
"code.gitea.io/gitea/modules/indexer/issues/internal"
|
|
|
|
|
|
|
|
"github.com/olivere/elastic/v7"
|
|
|
|
)
|
|
|
|
|
|
|
|
const (
|
|
|
|
issueIndexerLatestVersion = 0
|
|
|
|
)
|
|
|
|
|
|
|
|
var _ internal.Indexer = &Indexer{}
|
|
|
|
|
|
|
|
// Indexer implements Indexer interface
|
|
|
|
type Indexer struct {
|
|
|
|
inner *inner_elasticsearch.Indexer
|
|
|
|
indexer_internal.Indexer // do not composite inner_elasticsearch.Indexer directly to avoid exposing too much
|
|
|
|
}
|
|
|
|
|
|
|
|
// NewIndexer creates a new elasticsearch indexer
|
|
|
|
func NewIndexer(url, indexerName string) *Indexer {
|
|
|
|
inner := inner_elasticsearch.NewIndexer(url, indexerName, issueIndexerLatestVersion, defaultMapping)
|
|
|
|
indexer := &Indexer{
|
|
|
|
inner: inner,
|
|
|
|
Indexer: inner,
|
|
|
|
}
|
|
|
|
return indexer
|
|
|
|
}
|
|
|
|
|
|
|
|
const (
|
|
|
|
defaultMapping = `{
|
|
|
|
"mappings": {
|
|
|
|
"properties": {
|
|
|
|
"id": {
|
|
|
|
"type": "integer",
|
|
|
|
"index": true
|
|
|
|
},
|
|
|
|
"repo_id": {
|
|
|
|
"type": "integer",
|
|
|
|
"index": true
|
|
|
|
},
|
|
|
|
"title": {
|
|
|
|
"type": "text",
|
|
|
|
"index": true
|
|
|
|
},
|
|
|
|
"content": {
|
|
|
|
"type": "text",
|
|
|
|
"index": true
|
|
|
|
},
|
|
|
|
"comments": {
|
|
|
|
"type" : "text",
|
|
|
|
"index": true
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}`
|
|
|
|
)
|
|
|
|
|
|
|
|
// Index will save the index data
|
|
|
|
func (b *Indexer) Index(ctx context.Context, issues []*internal.IndexerData) error {
|
|
|
|
if len(issues) == 0 {
|
|
|
|
return nil
|
|
|
|
} else if len(issues) == 1 {
|
|
|
|
issue := issues[0]
|
|
|
|
_, err := b.inner.Client.Index().
|
|
|
|
Index(b.inner.VersionedIndexName()).
|
|
|
|
Id(fmt.Sprintf("%d", issue.ID)).
|
2023-07-04 14:36:08 -04:00
|
|
|
BodyJson(map[string]any{
|
Refactor indexer (#25174)
Refactor `modules/indexer` to make it more maintainable. And it can be
easier to support more features. I'm trying to solve some of issue
searching, this is a precursor to making functional changes.
Current supported engines and the index versions:
| engines | issues | code |
| - | - | - |
| db | Just a wrapper for database queries, doesn't need version | - |
| bleve | The version of index is **2** | The version of index is **6**
|
| elasticsearch | The old index has no version, will be treated as
version **0** in this PR | The version of index is **1** |
| meilisearch | The old index has no version, will be treated as version
**0** in this PR | - |
## Changes
### Split
Splited it into mutiple packages
```text
indexer
├── internal
│ ├── bleve
│ ├── db
│ ├── elasticsearch
│ └── meilisearch
├── code
│ ├── bleve
│ ├── elasticsearch
│ └── internal
└── issues
├── bleve
├── db
├── elasticsearch
├── internal
└── meilisearch
```
- `indexer/interanal`: Internal shared package for indexer.
- `indexer/interanal/[engine]`: Internal shared package for each engine
(bleve/db/elasticsearch/meilisearch).
- `indexer/code`: Implementations for code indexer.
- `indexer/code/internal`: Internal shared package for code indexer.
- `indexer/code/[engine]`: Implementation via each engine for code
indexer.
- `indexer/issues`: Implementations for issues indexer.
### Deduplication
- Combine `Init/Ping/Close` for code indexer and issues indexer.
- ~Combine `issues.indexerHolder` and `code.wrappedIndexer` to
`internal.IndexHolder`.~ Remove it, use dummy indexer instead when the
indexer is not ready.
- Duplicate two copies of creating ES clients.
- Duplicate two copies of `indexerID()`.
### Enhancement
- [x] Support index version for elasticsearch issues indexer, the old
index without version will be treated as version 0.
- [x] Fix spell of `elastic_search/ElasticSearch`, it should be
`Elasticsearch`.
- [x] Improve versioning of ES index. We don't need `Aliases`:
- Gitea does't need aliases for "Zero Downtime" because it never delete
old indexes.
- The old code of issues indexer uses the orignal name to create issue
index, so it's tricky to convert it to an alias.
- [x] Support index version for meilisearch issues indexer, the old
index without version will be treated as version 0.
- [x] Do "ping" only when `Ping` has been called, don't ping
periodically and cache the status.
- [x] Support the context parameter whenever possible.
- [x] Fix outdated example config.
- [x] Give up the requeue logic of issues indexer: When indexing fails,
call Ping to check if it was caused by the engine being unavailable, and
only requeue the task if the engine is unavailable.
- It is fragile and tricky, could cause data losing (It did happen when
I was doing some tests for this PR). And it works for ES only.
- Just always requeue the failed task, if it caused by bad data, it's a
bug of Gitea which should be fixed.
---------
Co-authored-by: Giteabot <teabot@gitea.io>
2023-06-23 08:37:56 -04:00
|
|
|
"id": issue.ID,
|
|
|
|
"repo_id": issue.RepoID,
|
|
|
|
"title": issue.Title,
|
|
|
|
"content": issue.Content,
|
|
|
|
"comments": issue.Comments,
|
|
|
|
}).
|
|
|
|
Do(ctx)
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
|
|
|
|
reqs := make([]elastic.BulkableRequest, 0)
|
|
|
|
for _, issue := range issues {
|
|
|
|
reqs = append(reqs,
|
|
|
|
elastic.NewBulkIndexRequest().
|
|
|
|
Index(b.inner.VersionedIndexName()).
|
|
|
|
Id(fmt.Sprintf("%d", issue.ID)).
|
2023-07-04 14:36:08 -04:00
|
|
|
Doc(map[string]any{
|
Refactor indexer (#25174)
Refactor `modules/indexer` to make it more maintainable. And it can be
easier to support more features. I'm trying to solve some of issue
searching, this is a precursor to making functional changes.
Current supported engines and the index versions:
| engines | issues | code |
| - | - | - |
| db | Just a wrapper for database queries, doesn't need version | - |
| bleve | The version of index is **2** | The version of index is **6**
|
| elasticsearch | The old index has no version, will be treated as
version **0** in this PR | The version of index is **1** |
| meilisearch | The old index has no version, will be treated as version
**0** in this PR | - |
## Changes
### Split
Splited it into mutiple packages
```text
indexer
├── internal
│ ├── bleve
│ ├── db
│ ├── elasticsearch
│ └── meilisearch
├── code
│ ├── bleve
│ ├── elasticsearch
│ └── internal
└── issues
├── bleve
├── db
├── elasticsearch
├── internal
└── meilisearch
```
- `indexer/interanal`: Internal shared package for indexer.
- `indexer/interanal/[engine]`: Internal shared package for each engine
(bleve/db/elasticsearch/meilisearch).
- `indexer/code`: Implementations for code indexer.
- `indexer/code/internal`: Internal shared package for code indexer.
- `indexer/code/[engine]`: Implementation via each engine for code
indexer.
- `indexer/issues`: Implementations for issues indexer.
### Deduplication
- Combine `Init/Ping/Close` for code indexer and issues indexer.
- ~Combine `issues.indexerHolder` and `code.wrappedIndexer` to
`internal.IndexHolder`.~ Remove it, use dummy indexer instead when the
indexer is not ready.
- Duplicate two copies of creating ES clients.
- Duplicate two copies of `indexerID()`.
### Enhancement
- [x] Support index version for elasticsearch issues indexer, the old
index without version will be treated as version 0.
- [x] Fix spell of `elastic_search/ElasticSearch`, it should be
`Elasticsearch`.
- [x] Improve versioning of ES index. We don't need `Aliases`:
- Gitea does't need aliases for "Zero Downtime" because it never delete
old indexes.
- The old code of issues indexer uses the orignal name to create issue
index, so it's tricky to convert it to an alias.
- [x] Support index version for meilisearch issues indexer, the old
index without version will be treated as version 0.
- [x] Do "ping" only when `Ping` has been called, don't ping
periodically and cache the status.
- [x] Support the context parameter whenever possible.
- [x] Fix outdated example config.
- [x] Give up the requeue logic of issues indexer: When indexing fails,
call Ping to check if it was caused by the engine being unavailable, and
only requeue the task if the engine is unavailable.
- It is fragile and tricky, could cause data losing (It did happen when
I was doing some tests for this PR). And it works for ES only.
- Just always requeue the failed task, if it caused by bad data, it's a
bug of Gitea which should be fixed.
---------
Co-authored-by: Giteabot <teabot@gitea.io>
2023-06-23 08:37:56 -04:00
|
|
|
"id": issue.ID,
|
|
|
|
"repo_id": issue.RepoID,
|
|
|
|
"title": issue.Title,
|
|
|
|
"content": issue.Content,
|
|
|
|
"comments": issue.Comments,
|
|
|
|
}),
|
|
|
|
)
|
|
|
|
}
|
|
|
|
|
|
|
|
_, err := b.inner.Client.Bulk().
|
|
|
|
Index(b.inner.VersionedIndexName()).
|
|
|
|
Add(reqs...).
|
|
|
|
Do(graceful.GetManager().HammerContext())
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
|
|
|
|
// Delete deletes indexes by ids
|
|
|
|
func (b *Indexer) Delete(ctx context.Context, ids ...int64) error {
|
|
|
|
if len(ids) == 0 {
|
|
|
|
return nil
|
|
|
|
} else if len(ids) == 1 {
|
|
|
|
_, err := b.inner.Client.Delete().
|
|
|
|
Index(b.inner.VersionedIndexName()).
|
|
|
|
Id(fmt.Sprintf("%d", ids[0])).
|
|
|
|
Do(ctx)
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
|
|
|
|
reqs := make([]elastic.BulkableRequest, 0)
|
|
|
|
for _, id := range ids {
|
|
|
|
reqs = append(reqs,
|
|
|
|
elastic.NewBulkDeleteRequest().
|
|
|
|
Index(b.inner.VersionedIndexName()).
|
|
|
|
Id(fmt.Sprintf("%d", id)),
|
|
|
|
)
|
|
|
|
}
|
|
|
|
|
|
|
|
_, err := b.inner.Client.Bulk().
|
|
|
|
Index(b.inner.VersionedIndexName()).
|
|
|
|
Add(reqs...).
|
|
|
|
Do(graceful.GetManager().HammerContext())
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
|
|
|
|
// Search searches for issues by given conditions.
|
|
|
|
// Returns the matching issue IDs
|
|
|
|
func (b *Indexer) Search(ctx context.Context, keyword string, repoIDs []int64, limit, start int) (*internal.SearchResult, error) {
|
|
|
|
kwQuery := elastic.NewMultiMatchQuery(keyword, "title", "content", "comments")
|
|
|
|
query := elastic.NewBoolQuery()
|
|
|
|
query = query.Must(kwQuery)
|
|
|
|
if len(repoIDs) > 0 {
|
2023-07-04 14:36:08 -04:00
|
|
|
repoStrs := make([]any, 0, len(repoIDs))
|
Refactor indexer (#25174)
Refactor `modules/indexer` to make it more maintainable. And it can be
easier to support more features. I'm trying to solve some of issue
searching, this is a precursor to making functional changes.
Current supported engines and the index versions:
| engines | issues | code |
| - | - | - |
| db | Just a wrapper for database queries, doesn't need version | - |
| bleve | The version of index is **2** | The version of index is **6**
|
| elasticsearch | The old index has no version, will be treated as
version **0** in this PR | The version of index is **1** |
| meilisearch | The old index has no version, will be treated as version
**0** in this PR | - |
## Changes
### Split
Splited it into mutiple packages
```text
indexer
├── internal
│ ├── bleve
│ ├── db
│ ├── elasticsearch
│ └── meilisearch
├── code
│ ├── bleve
│ ├── elasticsearch
│ └── internal
└── issues
├── bleve
├── db
├── elasticsearch
├── internal
└── meilisearch
```
- `indexer/interanal`: Internal shared package for indexer.
- `indexer/interanal/[engine]`: Internal shared package for each engine
(bleve/db/elasticsearch/meilisearch).
- `indexer/code`: Implementations for code indexer.
- `indexer/code/internal`: Internal shared package for code indexer.
- `indexer/code/[engine]`: Implementation via each engine for code
indexer.
- `indexer/issues`: Implementations for issues indexer.
### Deduplication
- Combine `Init/Ping/Close` for code indexer and issues indexer.
- ~Combine `issues.indexerHolder` and `code.wrappedIndexer` to
`internal.IndexHolder`.~ Remove it, use dummy indexer instead when the
indexer is not ready.
- Duplicate two copies of creating ES clients.
- Duplicate two copies of `indexerID()`.
### Enhancement
- [x] Support index version for elasticsearch issues indexer, the old
index without version will be treated as version 0.
- [x] Fix spell of `elastic_search/ElasticSearch`, it should be
`Elasticsearch`.
- [x] Improve versioning of ES index. We don't need `Aliases`:
- Gitea does't need aliases for "Zero Downtime" because it never delete
old indexes.
- The old code of issues indexer uses the orignal name to create issue
index, so it's tricky to convert it to an alias.
- [x] Support index version for meilisearch issues indexer, the old
index without version will be treated as version 0.
- [x] Do "ping" only when `Ping` has been called, don't ping
periodically and cache the status.
- [x] Support the context parameter whenever possible.
- [x] Fix outdated example config.
- [x] Give up the requeue logic of issues indexer: When indexing fails,
call Ping to check if it was caused by the engine being unavailable, and
only requeue the task if the engine is unavailable.
- It is fragile and tricky, could cause data losing (It did happen when
I was doing some tests for this PR). And it works for ES only.
- Just always requeue the failed task, if it caused by bad data, it's a
bug of Gitea which should be fixed.
---------
Co-authored-by: Giteabot <teabot@gitea.io>
2023-06-23 08:37:56 -04:00
|
|
|
for _, repoID := range repoIDs {
|
|
|
|
repoStrs = append(repoStrs, repoID)
|
|
|
|
}
|
|
|
|
repoQuery := elastic.NewTermsQuery("repo_id", repoStrs...)
|
|
|
|
query = query.Must(repoQuery)
|
|
|
|
}
|
|
|
|
searchResult, err := b.inner.Client.Search().
|
|
|
|
Index(b.inner.VersionedIndexName()).
|
|
|
|
Query(query).
|
|
|
|
Sort("_score", false).
|
|
|
|
From(start).Size(limit).
|
|
|
|
Do(ctx)
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
|
|
|
|
hits := make([]internal.Match, 0, limit)
|
|
|
|
for _, hit := range searchResult.Hits.Hits {
|
|
|
|
id, _ := strconv.ParseInt(hit.Id, 10, 64)
|
|
|
|
hits = append(hits, internal.Match{
|
|
|
|
ID: id,
|
|
|
|
})
|
|
|
|
}
|
|
|
|
|
|
|
|
return &internal.SearchResult{
|
|
|
|
Total: searchResult.TotalHits(),
|
|
|
|
Hits: hits,
|
|
|
|
}, nil
|
|
|
|
}
|