1
1
mirror of https://github.com/go-gitea/gitea synced 2024-11-17 23:54:25 +00:00
gitea/modules/indexer/issues/internal/model.go
Jason Song 1e76a824bc
Refactor and enhance issue indexer to support both searching, filtering and paging (#26012)
Fix #24662.

Replace #24822 and #25708 (although it has been merged)


## Background

In the past, Gitea supported issue searching with a keyword and
conditions in a less efficient way. It worked by searching for issues
with the keyword and obtaining limited IDs (as it is heavy to get all)
on the indexer (bleve/elasticsearch/meilisearch), and then querying with
conditions on the database to find a subset of the found IDs. This is
why the results could be incomplete.

To solve this issue, we need to store all fields that could be used as
conditions in the indexer and support both keyword and additional
conditions when searching with the indexer.

## Major changes

- Redefine `IndexerData` to include all fields that could be used as
filter conditions.
- Refactor `Search(ctx context.Context, kw string, repoIDs []int64,
limit, start int, state string)` to `Search(ctx context.Context, options
*SearchOptions)`, so it supports more conditions now.
- Change the data type stored in `issueIndexerQueue`. Use
`IndexerMetadata` instead of `IndexerData` in case the data has been
updated while it is in the queue. This also reduces the storage size of
the queue.
- Enhance searching with Bleve/Elasticsearch/Meilisearch, make them
fully support `SearchOptions`. Also, update the data versions.
- Keep most logic of database indexer, but remove
`issues.SearchIssueIDsByKeyword` in `models` to avoid confusion where is
the entry point to search issues.
- Start a Meilisearch instance to test it in unit tests.
- Add unit tests with almost full coverage to test
Bleve/Elasticsearch/Meilisearch indexer.

---------

Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
2023-07-31 06:28:53 +00:00

122 lines
4.6 KiB
Go

// Copyright 2023 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package internal
import (
"code.gitea.io/gitea/models/db"
"code.gitea.io/gitea/modules/timeutil"
"code.gitea.io/gitea/modules/util"
)
// IndexerData data stored in the issue indexer
type IndexerData struct {
ID int64 `json:"id"`
RepoID int64 `json:"repo_id"`
IsPublic bool `json:"is_public"` // If the repo is public
// Fields used for keyword searching
Title string `json:"title"`
Content string `json:"content"`
Comments []string `json:"comments"`
// Fields used for filtering
IsPull bool `json:"is_pull"`
IsClosed bool `json:"is_closed"`
LabelIDs []int64 `json:"label_ids"`
NoLabel bool `json:"no_label"` // True if LabelIDs is empty
MilestoneID int64 `json:"milestone_id"`
ProjectID int64 `json:"project_id"`
ProjectBoardID int64 `json:"project_board_id"`
PosterID int64 `json:"poster_id"`
AssigneeID int64 `json:"assignee_id"`
MentionIDs []int64 `json:"mention_ids"`
ReviewedIDs []int64 `json:"reviewed_ids"`
ReviewRequestedIDs []int64 `json:"review_requested_ids"`
SubscriberIDs []int64 `json:"subscriber_ids"`
UpdatedUnix timeutil.TimeStamp `json:"updated_unix"`
// Fields used for sorting
// UpdatedUnix is both used for filtering and sorting.
// ID is used for sorting too, to make the sorting stable.
CreatedUnix timeutil.TimeStamp `json:"created_unix"`
DeadlineUnix timeutil.TimeStamp `json:"deadline_unix"`
CommentCount int64 `json:"comment_count"`
}
// Match represents on search result
type Match struct {
ID int64 `json:"id"`
Score float64 `json:"score"`
}
// SearchResult represents search results
type SearchResult struct {
Total int64
Hits []Match
}
// SearchOptions represents search options
type SearchOptions struct {
Keyword string // keyword to search
RepoIDs []int64 // repository IDs which the issues belong to
AllPublic bool // if include all public repositories
IsPull util.OptionalBool // if the issues is a pull request
IsClosed util.OptionalBool // if the issues is closed
IncludedLabelIDs []int64 // labels the issues have
ExcludedLabelIDs []int64 // labels the issues don't have
IncludedAnyLabelIDs []int64 // labels the issues have at least one. It will be ignored if IncludedLabelIDs is not empty. It's an uncommon filter, but it has been supported accidentally by issues.IssuesOptions.IncludedLabelNames.
NoLabelOnly bool // if the issues have no label, if true, IncludedLabelIDs and ExcludedLabelIDs, IncludedAnyLabelIDs will be ignored
MilestoneIDs []int64 // milestones the issues have
ProjectID *int64 // project the issues belong to
ProjectBoardID *int64 // project board the issues belong to
PosterID *int64 // poster of the issues
AssigneeID *int64 // assignee of the issues, zero means no assignee
MentionID *int64 // mentioned user of the issues
ReviewedID *int64 // reviewer of the issues
ReviewRequestedID *int64 // requested reviewer of the issues
SubscriberID *int64 // subscriber of the issues
UpdatedAfterUnix *int64
UpdatedBeforeUnix *int64
db.Paginator
SortBy SortBy // sort by field
}
type SortBy string
const (
SortByCreatedDesc SortBy = "-created_unix"
SortByUpdatedDesc SortBy = "-updated_unix"
SortByCommentsDesc SortBy = "-comment_count"
SortByDeadlineDesc SortBy = "-deadline_unix"
SortByCreatedAsc SortBy = "created_unix"
SortByUpdatedAsc SortBy = "updated_unix"
SortByCommentsAsc SortBy = "comment_count"
SortByDeadlineAsc SortBy = "deadline_unix"
// Unsupported sort types which are supported by issues.IssuesOptions.SortType:
//
// - "priorityrepo":
// It's impossible to support it in the indexer.
// It is based on the specified repository in the request, so we cannot add static field to the indexer.
// If we do something like that query the issues in the specified repository first then append other issues,
// it will break the pagination.
//
// - "project-column-sorting":
// Although it's possible to support it by adding project.ProjectIssue.Sorting to the indexer,
// but what if the issue belongs to multiple projects?
// Since it's unsupported to search issues with keyword in project page, we don't need to support it.
)