MirrorHub/gitea
0
0
Fork 0
mirror of https://github.com/go-gitea/gitea synced 2024-11-28 07:12:45 +01:00
Code Issues Releases Wiki Activity
gitea/models/repo/repo_list_test.go
Bruno Sofiato 900ac62251
Allow code search by filename (#32210) 
This is a large and complex PR, so let me explain in detail its changes.

First, I had to create new index mappings for Bleve and ElasticSerach as
the current ones do not support search by filename. This requires Gitea
to recreate the code search indexes (I do not know if this is a breaking
change, but I feel it deserves a heads-up).

I've used [this
approach](https://www.elastic.co/guide/en/elasticsearch/reference/7.17/analysis-pathhierarchy-tokenizer.html)
to model the filename index. It allows us to efficiently search for both
the full path and the name of a file. Bleve, however, does not support
this out-of-box, so I had to code a brand new [token
filter](https://blevesearch.com/docs/Token-Filters/) to generate the
search terms.

I also did an overhaul in the `indexer_test.go` file. It now asserts the
order of the expected results (this is important since matches based on
the name of a file are more relevant than those based on its content).
I've added new test scenarios that deal with searching by filename. They
use a new repo included in the Gitea fixture.

The screenshot below depicts how Gitea shows the search results. It
shows results based on content in the same way as the current version
does. In matches based on the filename, the first seven lines of the
file contents are shown (BTW, this is how GitHub does it).


![image](https://github.com/user-attachments/assets/9d938d86-1a8d-4f89-8644-1921a473e858)

Resolves #32096

---------

Signed-off-by: Bruno Sofiato <bruno.sofiato@gmail.com>
2024-10-11 23:35:04 +00:00

404 lines
14 KiB
Go
Raw Blame History

// Copyright 2017 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package repo_test
import (
"strings"
"testing"
"code.gitea.io/gitea/models/db"
repo_model "code.gitea.io/gitea/models/repo"
"code.gitea.io/gitea/models/unittest"
"code.gitea.io/gitea/modules/optional"
"github.com/stretchr/testify/assert"
)
func getTestCases() []struct {
name string
opts *repo_model.SearchRepoOptions
count int
} {
testCases := []struct {
name string
opts *repo_model.SearchRepoOptions
count int
}{
{
name: "PublicRepositoriesByName",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{PageSize: 10}, Collaborate: optional.Some(false)},
count: 7,
},
{
name: "PublicAndPrivateRepositoriesByName",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{Page: 1, PageSize: 10}, Private: true, Collaborate: optional.Some(false)},
count: 14,
},
{
name: "PublicAndPrivateRepositoriesByNameWithPagesizeLimitFirstPage",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{Page: 1, PageSize: 5}, Private: true, Collaborate: optional.Some(false)},
count: 14,
},
{
name: "PublicAndPrivateRepositoriesByNameWithPagesizeLimitSecondPage",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{Page: 2, PageSize: 5}, Private: true, Collaborate: optional.Some(false)},
count: 14,
},
{
name: "PublicAndPrivateRepositoriesByNameWithPagesizeLimitThirdPage",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{Page: 3, PageSize: 5}, Private: true, Collaborate: optional.Some(false)},
count: 14,
},
{
name: "PublicAndPrivateRepositoriesByNameWithPagesizeLimitFourthPage",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{Page: 3, PageSize: 5}, Private: true, Collaborate: optional.Some(false)},
count: 14,
},
{
name: "PublicRepositoriesOfUser",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 15, Collaborate: optional.Some(false)},
count: 2,
},
{
name: "PublicRepositoriesOfUser2",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 18, Collaborate: optional.Some(false)},
count: 0,
},
{
name: "PublicRepositoriesOfOrg3",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 20, Collaborate: optional.Some(false)},
count: 2,
},
{
name: "PublicAndPrivateRepositoriesOfUser",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 15, Private: true, Collaborate: optional.Some(false)},
count: 4,
},
{
name: "PublicAndPrivateRepositoriesOfUser2",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 18, Private: true, Collaborate: optional.Some(false)},
count: 0,
},
{
name: "PublicAndPrivateRepositoriesOfOrg3",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 20, Private: true, Collaborate: optional.Some(false)},
count: 4,
},
{
name: "PublicRepositoriesOfUserIncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 15},
count: 5,
},
{
name: "PublicRepositoriesOfUser2IncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 18},
count: 1,
},
{
name: "PublicRepositoriesOfOrg3IncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 20},
count: 3,
},
{
name: "PublicAndPrivateRepositoriesOfUserIncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 15, Private: true},
count: 9,
},
{
name: "PublicAndPrivateRepositoriesOfUser2IncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 18, Private: true},
count: 4,
},
{
name: "PublicAndPrivateRepositoriesOfOrg3IncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 20, Private: true},
count: 7,
},
{
name: "PublicRepositoriesOfOrganization",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 17, Collaborate: optional.Some(false)},
count: 1,
},
{
name: "PublicAndPrivateRepositoriesOfOrganization",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 17, Private: true, Collaborate: optional.Some(false)},
count: 2,
},
{
name: "AllPublic/PublicRepositoriesByName",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{PageSize: 10}, AllPublic: true, Collaborate: optional.Some(false)},
count: 7,
},
{
name: "AllPublic/PublicAndPrivateRepositoriesByName",
opts: &repo_model.SearchRepoOptions{Keyword: "big_test_", ListOptions: db.ListOptions{Page: 1, PageSize: 10}, Private: true, AllPublic: true, Collaborate: optional.Some(false)},
count: 14,
},
{
name: "AllPublic/PublicRepositoriesOfUserIncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 15, AllPublic: true, Template: optional.Some(false)},
count: 34,
},
{
name: "AllPublic/PublicAndPrivateRepositoriesOfUserIncludingCollaborative",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 15, Private: true, AllPublic: true, AllLimited: true, Template: optional.Some(false)},
count: 39,
},
{
name: "AllPublic/PublicAndPrivateRepositoriesOfUserIncludingCollaborativeByName",
opts: &repo_model.SearchRepoOptions{Keyword: "test", ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 15, Private: true, AllPublic: true},
count: 15,
},
{
name: "AllPublic/PublicAndPrivateRepositoriesOfUser2IncludingCollaborativeByName",
opts: &repo_model.SearchRepoOptions{Keyword: "test", ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 18, Private: true, AllPublic: true},
count: 13,
},
{
name: "AllPublic/PublicRepositoriesOfOrganization",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, OwnerID: 17, AllPublic: true, Collaborate: optional.Some(false), Template: optional.Some(false)},
count: 34,
},
{
name: "AllTemplates",
opts: &repo_model.SearchRepoOptions{ListOptions: db.ListOptions{Page: 1, PageSize: 10}, Template: optional.Some(true)},
count: 2,
},
{
name: "OwnerSlashRepoSearch",
opts: &repo_model.SearchRepoOptions{Keyword: "user/repo2", ListOptions: db.ListOptions{Page: 1, PageSize: 10}, Private: true, OwnerID: 0},
count: 2,
},
{
name: "OwnerSlashSearch",
opts: &repo_model.SearchRepoOptions{Keyword: "user20/", ListOptions: db.ListOptions{Page: 1, PageSize: 10}, Private: true, OwnerID: 0},
count: 4,
},
}
return testCases
}
func TestSearchRepository(t *testing.T) {
assert.NoError(t, unittest.PrepareTestDatabase())
// test search public repository on explore page
repos, count, err := repo_model.SearchRepositoryByName(db.DefaultContext, &repo_model.SearchRepoOptions{
ListOptions: db.ListOptions{
Page: 1,
PageSize: 10,
},
Keyword: "repo_12",
Collaborate: optional.Some(false),
})
assert.NoError(t, err)
if assert.Len(t, repos, 1) {
assert.Equal(t, "test_repo_12", repos[0].Name)
}
assert.Equal(t, int64(1), count)
repos, count, err = repo_model.SearchRepositoryByName(db.DefaultContext, &repo_model.SearchRepoOptions{
ListOptions: db.ListOptions{
Page: 1,
PageSize: 10,
},
Keyword: "test_repo",
Collaborate: optional.Some(false),
})
assert.NoError(t, err)
assert.Equal(t, int64(2), count)
assert.Len(t, repos, 2)
// test search private repository on explore page
repos, count, err = repo_model.SearchRepositoryByName(db.DefaultContext, &repo_model.SearchRepoOptions{
ListOptions: db.ListOptions{
Page: 1,
PageSize: 10,
},
Keyword: "repo_13",
Private: true,
Collaborate: optional.Some(false),
})
assert.NoError(t, err)
if assert.Len(t, repos, 1) {
assert.Equal(t, "test_repo_13", repos[0].Name)
}
assert.Equal(t, int64(1), count)
repos, count, err = repo_model.SearchRepositoryByName(db.DefaultContext, &repo_model.SearchRepoOptions{
ListOptions: db.ListOptions{
Page: 1,
PageSize: 10,
},
Keyword: "test_repo",
Private: true,
Collaborate: optional.Some(false),
})
assert.NoError(t, err)
assert.Equal(t, int64(3), count)
assert.Len(t, repos, 3)
// Test non existing owner
repos, count, err = repo_model.SearchRepositoryByName(db.DefaultContext, &repo_model.SearchRepoOptions{OwnerID: unittest.NonexistentID})
assert.NoError(t, err)
assert.Empty(t, repos)
assert.Equal(t, int64(0), count)
// Test search within description
repos, count, err = repo_model.SearchRepository(db.DefaultContext, &repo_model.SearchRepoOptions{
ListOptions: db.ListOptions{
Page: 1,
PageSize: 10,
},
Keyword: "description_14",
Collaborate: optional.Some(false),
IncludeDescription: true,
})
assert.NoError(t, err)
if assert.Len(t, repos, 1) {
assert.Equal(t, "test_repo_14", repos[0].Name)
}
assert.Equal(t, int64(1), count)
// Test NOT search within description
repos, count, err = repo_model.SearchRepository(db.DefaultContext, &repo_model.SearchRepoOptions{
ListOptions: db.ListOptions{
Page: 1,
PageSize: 10,
},
Keyword: "description_14",
Collaborate: optional.Some(false),
IncludeDescription: false,
})
assert.NoError(t, err)
assert.Empty(t, repos)
assert.Equal(t, int64(0), count)
testCases := getTestCases()
for _, testCase := range testCases {
t.Run(testCase.name, func(t *testing.T) {
repos, count, err := repo_model.SearchRepositoryByName(db.DefaultContext, testCase.opts)
assert.NoError(t, err)
assert.Equal(t, int64(testCase.count), count)
page := testCase.opts.Page
if page <= 0 {
page = 1
}
expectedLen := testCase.opts.PageSize
if testCase.opts.PageSize*page > testCase.count+testCase.opts.PageSize {
expectedLen = 0
} else if testCase.opts.PageSize*page > testCase.count {
expectedLen = testCase.count % testCase.opts.PageSize
}
if assert.Len(t, repos, expectedLen) {
for _, repo := range repos {
assert.NotEmpty(t, repo.Name)
if len(testCase.opts.Keyword) > 0 {
// Keyword match condition is different for search terms of form "owner/repo"
if strings.Count(testCase.opts.Keyword, "/") == 1 {
// May still match as a whole...
wholeMatch := strings.Contains(repo.Name, testCase.opts.Keyword)
pieces := strings.Split(testCase.opts.Keyword, "/")
ownerName := pieces[0]
repoName := pieces[1]
// ... or match in parts
splitMatch := strings.Contains(repo.OwnerName, ownerName) && strings.Contains(repo.Name, repoName)
assert.True(t, wholeMatch || splitMatch, "Keyword '%s' does not match repo '%s/%s'", testCase.opts.Keyword, repo.Owner.Name, repo.Name)
} else {
assert.Contains(t, repo.Name, testCase.opts.Keyword)
}
}
if !testCase.opts.Private {
assert.False(t, repo.IsPrivate)
}
if testCase.opts.Fork.Value() && testCase.opts.Mirror.Value() {
assert.True(t, repo.IsFork && repo.IsMirror)
} else {
if testCase.opts.Fork.Has() {
assert.Equal(t, testCase.opts.Fork.Value(), repo.IsFork)
}
if testCase.opts.Mirror.Has() {
assert.Equal(t, testCase.opts.Mirror.Value(), repo.IsMirror)
}
}
if testCase.opts.OwnerID > 0 && !testCase.opts.AllPublic {
if testCase.opts.Collaborate.Has() {
if testCase.opts.Collaborate.Value() {
assert.NotEqual(t, testCase.opts.OwnerID, repo.Owner.ID)
} else {
assert.Equal(t, testCase.opts.OwnerID, repo.Owner.ID)
}
}
}
}
}
})
}
}
func TestCountRepository(t *testing.T) {
assert.NoError(t, unittest.PrepareTestDatabase())
testCases := getTestCases()
for _, testCase := range testCases {
t.Run(testCase.name, func(t *testing.T) {
count, err := repo_model.CountRepository(db.DefaultContext, testCase.opts)
assert.NoError(t, err)
assert.Equal(t, int64(testCase.count), count)
})
}
}
func TestSearchRepositoryByTopicName(t *testing.T) {
assert.NoError(t, unittest.PrepareTestDatabase())
testCases := []struct {
name string
opts *repo_model.SearchRepoOptions
count int
}{
{
name: "AllPublic/SearchPublicRepositoriesFromTopicAndName",
opts: &repo_model.SearchRepoOptions{OwnerID: 21, AllPublic: true, Keyword: "graphql"},
count: 2,
},
{
name: "AllPublic/OnlySearchPublicRepositoriesFromTopic",
opts: &repo_model.SearchRepoOptions{OwnerID: 21, AllPublic: true, Keyword: "graphql", TopicOnly: true},
count: 1,
},
{
name: "AllPublic/OnlySearchMultipleKeywordPublicRepositoriesFromTopic",
opts: &repo_model.SearchRepoOptions{OwnerID: 21, AllPublic: true, Keyword: "graphql,golang", TopicOnly: true},
count: 2,
},
}
for _, testCase := range testCases {
t.Run(testCase.name, func(t *testing.T) {
_, count, err := repo_model.SearchRepositoryByName(db.DefaultContext, testCase.opts)
assert.NoError(t, err)
assert.Equal(t, int64(testCase.count), count)
})
}
}
Reference in a new issue View git blame Copy permalink