forgejo/modules/repository/repo.go

523 lines
16 KiB
Go
Raw Normal View History

// Copyright 2019 The Gitea Authors. All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
import (
"context"
"fmt"
"io"
"net/http"
"path"
"strings"
"time"
"code.gitea.io/gitea/models/db"
git_model "code.gitea.io/gitea/models/git"
"code.gitea.io/gitea/models/organization"
repo_model "code.gitea.io/gitea/models/repo"
user_model "code.gitea.io/gitea/models/user"
"code.gitea.io/gitea/modules/git"
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
"code.gitea.io/gitea/modules/lfs"
"code.gitea.io/gitea/modules/log"
"code.gitea.io/gitea/modules/migration"
"code.gitea.io/gitea/modules/setting"
"code.gitea.io/gitea/modules/timeutil"
"code.gitea.io/gitea/modules/util"
"gopkg.in/ini.v1"
)
/*
GitHub, GitLab, Gogs: *.wiki.git
BitBucket: *.git/wiki
*/
var commonWikiURLSuffixes = []string{".wiki.git", ".git/wiki"}
// WikiRemoteURL returns accessible repository URL for wiki if exists.
// Otherwise, it returns an empty string.
func WikiRemoteURL(ctx context.Context, remote string) string {
remote = strings.TrimSuffix(remote, ".git")
for _, suffix := range commonWikiURLSuffixes {
wikiURL := remote + suffix
if git.IsRepoURLAccessible(ctx, wikiURL) {
return wikiURL
}
}
return ""
}
// MigrateRepositoryGitData starts migrating git related data after created migrating repository
func MigrateRepositoryGitData(ctx context.Context, u *user_model.User,
repo *repo_model.Repository, opts migration.MigrateOptions,
httpTransport *http.Transport,
) (*repo_model.Repository, error) {
repoPath := repo_model.RepoPath(u.Name, opts.RepoName)
if u.IsOrganization() {
t, err := organization.OrgFromUser(u).GetOwnerTeam()
if err != nil {
return nil, err
}
repo.NumWatches = t.NumMembers
} else {
repo.NumWatches = 1
}
migrateTimeout := time.Duration(setting.Git.Timeout.Migrate) * time.Second
var err error
if err = util.RemoveAll(repoPath); err != nil {
return repo, fmt.Errorf("Failed to remove %s: %v", repoPath, err)
}
if err = git.Clone(ctx, opts.CloneAddr, repoPath, git.CloneRepoOptions{
Mirror: true,
Quiet: true,
Timeout: migrateTimeout,
SkipTLSVerify: setting.Migrations.SkipTLSVerify,
}); err != nil {
return repo, fmt.Errorf("Clone: %v", err)
}
if err := git.WriteCommitGraph(ctx, repoPath); err != nil {
return repo, err
}
if opts.Wiki {
wikiPath := repo_model.WikiPath(u.Name, opts.RepoName)
wikiRemotePath := WikiRemoteURL(ctx, opts.CloneAddr)
if len(wikiRemotePath) > 0 {
if err := util.RemoveAll(wikiPath); err != nil {
return repo, fmt.Errorf("Failed to remove %s: %v", wikiPath, err)
}
2022-04-30 14:50:56 +02:00
if err := git.Clone(ctx, wikiRemotePath, wikiPath, git.CloneRepoOptions{
Mirror: true,
Quiet: true,
Timeout: migrateTimeout,
Branch: "master",
SkipTLSVerify: setting.Migrations.SkipTLSVerify,
}); err != nil {
log.Warn("Clone wiki: %v", err)
if err := util.RemoveAll(wikiPath); err != nil {
return repo, fmt.Errorf("Failed to remove %s: %v", wikiPath, err)
}
2022-04-30 14:50:56 +02:00
} else {
if err := git.WriteCommitGraph(ctx, wikiPath); err != nil {
return repo, err
}
}
}
}
if repo.OwnerID == u.ID {
repo.Owner = u
}
if err = CheckDaemonExportOK(ctx, repo); err != nil {
return repo, fmt.Errorf("checkDaemonExportOK: %v", err)
}
if stdout, _, err := git.NewCommand(ctx, "update-server-info").
SetDescription(fmt.Sprintf("MigrateRepositoryGitData(git update-server-info): %s", repoPath)).
RunStdString(&git.RunOpts{Dir: repoPath}); err != nil {
log.Error("MigrateRepositoryGitData(git update-server-info) in %v: Stdout: %s\nError: %v", repo, stdout, err)
return repo, fmt.Errorf("error in MigrateRepositoryGitData(git update-server-info): %v", err)
}
gitRepo, err := git.OpenRepository(ctx, repoPath)
if err != nil {
return repo, fmt.Errorf("OpenRepository: %v", err)
}
defer gitRepo.Close()
repo.IsEmpty, err = gitRepo.IsEmpty()
if err != nil {
return repo, fmt.Errorf("git.IsEmpty: %v", err)
}
if !repo.IsEmpty {
if len(repo.DefaultBranch) == 0 {
// Try to get HEAD branch and set it as default branch.
headBranch, err := gitRepo.GetHEADBranch()
if err != nil {
return repo, fmt.Errorf("GetHEADBranch: %v", err)
}
if headBranch != nil {
repo.DefaultBranch = headBranch.Name
}
}
if !opts.Releases {
Improve sync performance for pull-mirrors (#19125) This addresses https://github.com/go-gitea/gitea/issues/18352 It aims to improve performance (and resource use) of the `SyncReleasesWithTags` operation for pull-mirrors. For large repositories with many tags, `SyncReleasesWithTags` can be a costly operation (taking several minutes to complete). The reason is two-fold: 1. on sync, every upstream repo tag is compared (for changes) against existing local entries in the release table to ensure that they are up-to-date. 2. the procedure for getting _each tag_ involves a series of git operations ```bash git show-ref --tags -- v8.2.4477 git cat-file -t 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git cat-file -p 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git rev-list --count 29ab6ce9f36660cffaad3c8789e71162e5db5d2f ``` of which the `git rev-list --count` can be particularly heavy. This PR optimizes performance for pull-mirrors. We utilize the fact that a pull-mirror is always identical to its upstream and rebuild the entire release table on every sync and use a batch `git for-each-ref .. refs/tags` call to retrieve all tags in one go. For large mirror repos, with hundreds of annotated tags, this brings down the duration of the sync operation from several minutes to a few seconds. A few unscientific examples run on my local machine: - https://github.com/spring-projects/spring-boot (223 tags) - before: `0m28,673s` - after: `0m2,244s` - https://github.com/kubernetes/kubernetes (890 tags) - before: `8m00s` - after: `0m8,520s` - https://github.com/vim/vim (13954 tags) - before: `14m20,383s` - after: `0m35,467s` I added a `foreachref` package which contains a flexible way of specifying which reference fields are of interest (`git-for-each-ref(1)`) and to produce a parser for the expected output. These could be reused in other places where `for-each-ref` is used. I'll add unit tests for those if the overall PR looks promising.
2022-03-31 14:30:40 +02:00
// note: this will greatly improve release (tag) sync
// for pull-mirrors with many tags
repo.IsMirror = opts.Mirror
if err = SyncReleasesWithTags(repo, gitRepo); err != nil {
log.Error("Failed to synchronize tags to releases for repository: %v", err)
}
}
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
if opts.LFS {
endpoint := lfs.DetermineEndpoint(opts.CloneAddr, opts.LFSEndpoint)
lfsClient := lfs.NewClient(endpoint, httpTransport)
if err = StoreMissingLfsObjectsInRepository(ctx, repo, gitRepo, lfsClient); err != nil {
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
log.Error("Failed to store missing LFS objects for repository: %v", err)
}
}
}
ctx, committer, err := db.TxContext()
if err != nil {
return nil, err
}
defer committer.Close()
if opts.Mirror {
mirrorModel := repo_model.Mirror{
RepoID: repo.ID,
Interval: setting.Mirror.DefaultInterval,
EnablePrune: true,
NextUpdateUnix: timeutil.TimeStampNow().AddDuration(setting.Mirror.DefaultInterval),
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
LFS: opts.LFS,
}
if opts.LFS {
mirrorModel.LFSEndpoint = opts.LFSEndpoint
}
if opts.MirrorInterval != "" {
parsedInterval, err := time.ParseDuration(opts.MirrorInterval)
if err != nil {
log.Error("Failed to set Interval: %v", err)
return repo, err
}
if parsedInterval == 0 {
mirrorModel.Interval = 0
mirrorModel.NextUpdateUnix = 0
} else if parsedInterval < setting.Mirror.MinInterval {
err := fmt.Errorf("Interval %s is set below Minimum Interval of %s", parsedInterval, setting.Mirror.MinInterval)
log.Error("Interval: %s is too frequent", opts.MirrorInterval)
return repo, err
} else {
mirrorModel.Interval = parsedInterval
mirrorModel.NextUpdateUnix = timeutil.TimeStampNow().AddDuration(parsedInterval)
}
}
if err = repo_model.InsertMirror(ctx, &mirrorModel); err != nil {
return repo, fmt.Errorf("InsertOne: %v", err)
}
repo.IsMirror = true
if err = UpdateRepository(ctx, repo, false); err != nil {
return nil, err
}
} else {
if err = UpdateRepoSize(ctx, repo); err != nil {
log.Error("Failed to update size for repository: %v", err)
}
if repo, err = CleanUpMigrateInfo(ctx, repo); err != nil {
return nil, err
}
}
return repo, committer.Commit()
}
// cleanUpMigrateGitConfig removes mirror info which prevents "push --all".
// This also removes possible user credentials.
func cleanUpMigrateGitConfig(configPath string) error {
cfg, err := ini.Load(configPath)
if err != nil {
return fmt.Errorf("open config file: %v", err)
}
cfg.DeleteSection("remote \"origin\"")
if err = cfg.SaveToIndent(configPath, "\t"); err != nil {
return fmt.Errorf("save config file: %v", err)
}
return nil
}
// CleanUpMigrateInfo finishes migrating repository and/or wiki with things that don't need to be done for mirrors.
func CleanUpMigrateInfo(ctx context.Context, repo *repo_model.Repository) (*repo_model.Repository, error) {
repoPath := repo.RepoPath()
if err := createDelegateHooks(repoPath); err != nil {
return repo, fmt.Errorf("createDelegateHooks: %v", err)
}
if repo.HasWiki() {
if err := createDelegateHooks(repo.WikiPath()); err != nil {
return repo, fmt.Errorf("createDelegateHooks.(wiki): %v", err)
}
}
_, _, err := git.NewCommand(ctx, "remote", "rm", "origin").RunStdString(&git.RunOpts{Dir: repoPath})
if err != nil && !strings.HasPrefix(err.Error(), "exit status 128 - fatal: No such remote ") {
return repo, fmt.Errorf("CleanUpMigrateInfo: %v", err)
}
if repo.HasWiki() {
if err := cleanUpMigrateGitConfig(path.Join(repo.WikiPath(), "config")); err != nil {
return repo, fmt.Errorf("cleanUpMigrateGitConfig (wiki): %v", err)
}
}
return repo, UpdateRepository(ctx, repo, false)
}
// SyncReleasesWithTags synchronizes release table with repository tags
func SyncReleasesWithTags(repo *repo_model.Repository, gitRepo *git.Repository) error {
Improve sync performance for pull-mirrors (#19125) This addresses https://github.com/go-gitea/gitea/issues/18352 It aims to improve performance (and resource use) of the `SyncReleasesWithTags` operation for pull-mirrors. For large repositories with many tags, `SyncReleasesWithTags` can be a costly operation (taking several minutes to complete). The reason is two-fold: 1. on sync, every upstream repo tag is compared (for changes) against existing local entries in the release table to ensure that they are up-to-date. 2. the procedure for getting _each tag_ involves a series of git operations ```bash git show-ref --tags -- v8.2.4477 git cat-file -t 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git cat-file -p 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git rev-list --count 29ab6ce9f36660cffaad3c8789e71162e5db5d2f ``` of which the `git rev-list --count` can be particularly heavy. This PR optimizes performance for pull-mirrors. We utilize the fact that a pull-mirror is always identical to its upstream and rebuild the entire release table on every sync and use a batch `git for-each-ref .. refs/tags` call to retrieve all tags in one go. For large mirror repos, with hundreds of annotated tags, this brings down the duration of the sync operation from several minutes to a few seconds. A few unscientific examples run on my local machine: - https://github.com/spring-projects/spring-boot (223 tags) - before: `0m28,673s` - after: `0m2,244s` - https://github.com/kubernetes/kubernetes (890 tags) - before: `8m00s` - after: `0m8,520s` - https://github.com/vim/vim (13954 tags) - before: `14m20,383s` - after: `0m35,467s` I added a `foreachref` package which contains a flexible way of specifying which reference fields are of interest (`git-for-each-ref(1)`) and to produce a parser for the expected output. These could be reused in other places where `for-each-ref` is used. I'll add unit tests for those if the overall PR looks promising.
2022-03-31 14:30:40 +02:00
log.Debug("SyncReleasesWithTags: in Repo[%d:%s/%s]", repo.ID, repo.OwnerName, repo.Name)
// optimized procedure for pull-mirrors which saves a lot of time (in
// particular for repos with many tags).
if repo.IsMirror {
return pullMirrorReleaseSync(repo, gitRepo)
}
existingRelTags := make(map[string]struct{})
opts := repo_model.FindReleasesOptions{
IncludeDrafts: true,
IncludeTags: true,
ListOptions: db.ListOptions{PageSize: 50},
}
for page := 1; ; page++ {
API add/generalize pagination (#9452) * paginate results * fixed deadlock * prevented breaking change * updated swagger * go fmt * fixed find topic * go mod tidy * go mod vendor with go1.13.5 * fixed repo find topics * fixed unit test * added Limit method to Engine struct; use engine variable when provided; fixed gitignore * use ItemsPerPage for default pagesize; fix GetWatchers, getOrgUsersByOrgID and GetStargazers; fix GetAllCommits headers; reverted some changed behaviors * set Page value on Home route * improved memory allocations * fixed response headers * removed logfiles * fixed import order * import order * improved swagger * added function to get models.ListOptions from context * removed pagesize diff on unit test * fixed imports * removed unnecessary struct field * fixed go fmt * scoped PR * code improvements * code improvements * go mod tidy * fixed import order * fixed commit statuses session * fixed files headers * fixed headers; added pagination for notifications * go mod tidy * go fmt * removed Private from user search options; added setting.UI.IssuePagingNum as default valeu on repo's issues list * Apply suggestions from code review Co-Authored-By: 6543 <6543@obermui.de> Co-Authored-By: zeripath <art27@cantab.net> * fixed build error * CI.restart() * fixed merge conflicts resolve * fixed conflicts resolve * improved FindTrackedTimesOptions.ToOptions() method * added backwards compatibility on ListReleases request; fixed issue tracked time ToSession * fixed build error; fixed swagger template * fixed swagger template * fixed ListReleases backwards compatibility * added page to user search route Co-authored-by: techknowlogick <matti@mdranta.net> Co-authored-by: 6543 <6543@obermui.de> Co-authored-by: zeripath <art27@cantab.net>
2020-01-24 20:00:29 +01:00
opts.Page = page
rels, err := repo_model.GetReleasesByRepoID(repo.ID, opts)
if err != nil {
return fmt.Errorf("unable to GetReleasesByRepoID in Repo[%d:%s/%s]: %w", repo.ID, repo.OwnerName, repo.Name, err)
}
if len(rels) == 0 {
break
}
for _, rel := range rels {
if rel.IsDraft {
continue
}
commitID, err := gitRepo.GetTagCommitID(rel.TagName)
if err != nil && !git.IsErrNotExist(err) {
return fmt.Errorf("unable to GetTagCommitID for %q in Repo[%d:%s/%s]: %w", rel.TagName, repo.ID, repo.OwnerName, repo.Name, err)
}
if git.IsErrNotExist(err) || commitID != rel.Sha1 {
if err := repo_model.PushUpdateDeleteTag(repo, rel.TagName); err != nil {
return fmt.Errorf("unable to PushUpdateDeleteTag: %q in Repo[%d:%s/%s]: %w", rel.TagName, repo.ID, repo.OwnerName, repo.Name, err)
}
} else {
existingRelTags[strings.ToLower(rel.TagName)] = struct{}{}
}
}
}
_, err := gitRepo.WalkReferences(git.ObjectTag, 0, 0, func(sha1, refname string) error {
tagName := strings.TrimPrefix(refname, git.TagPrefix)
if _, ok := existingRelTags[strings.ToLower(tagName)]; ok {
return nil
}
if err := PushUpdateAddTag(repo, gitRepo, tagName, sha1, refname); err != nil {
return fmt.Errorf("unable to PushUpdateAddTag: %q to Repo[%d:%s/%s]: %w", tagName, repo.ID, repo.OwnerName, repo.Name, err)
}
return nil
})
return err
}
// PushUpdateAddTag must be called for any push actions to add tag
func PushUpdateAddTag(repo *repo_model.Repository, gitRepo *git.Repository, tagName, sha1, refname string) error {
tag, err := gitRepo.GetTagWithID(sha1, tagName)
if err != nil {
return fmt.Errorf("unable to GetTag: %w", err)
}
commit, err := tag.Commit(gitRepo)
if err != nil {
return fmt.Errorf("unable to get tag Commit: %w", err)
}
sig := tag.Tagger
if sig == nil {
sig = commit.Author
}
if sig == nil {
sig = commit.Committer
}
var author *user_model.User
createdAt := time.Unix(1, 0)
if sig != nil {
author, err = user_model.GetUserByEmail(sig.Email)
if err != nil && !user_model.IsErrUserNotExist(err) {
return fmt.Errorf("unable to GetUserByEmail for %q: %w", sig.Email, err)
}
createdAt = sig.When
}
commitsCount, err := commit.CommitsCount()
if err != nil {
return fmt.Errorf("unable to get CommitsCount: %w", err)
}
rel := repo_model.Release{
RepoID: repo.ID,
TagName: tagName,
LowerTagName: strings.ToLower(tagName),
Sha1: commit.ID.String(),
NumCommits: commitsCount,
CreatedUnix: timeutil.TimeStamp(createdAt.Unix()),
IsTag: true,
}
if author != nil {
rel.PublisherID = author.ID
}
return repo_model.SaveOrUpdateTag(repo, &rel)
}
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
// StoreMissingLfsObjectsInRepository downloads missing LFS objects
func StoreMissingLfsObjectsInRepository(ctx context.Context, repo *repo_model.Repository, gitRepo *git.Repository, lfsClient lfs.Client) error {
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
contentStore := lfs.NewContentStore()
pointerChan := make(chan lfs.PointerBlob)
errChan := make(chan error, 1)
go lfs.SearchPointerBlobs(ctx, gitRepo, pointerChan, errChan)
downloadObjects := func(pointers []lfs.Pointer) error {
err := lfsClient.Download(ctx, pointers, func(p lfs.Pointer, content io.ReadCloser, objectError error) error {
if objectError != nil {
return objectError
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
}
defer content.Close()
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
_, err := git_model.NewLFSMetaObject(&git_model.LFSMetaObject{Pointer: p, RepositoryID: repo.ID})
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
if err != nil {
log.Error("Repo[%-v]: Error creating LFS meta object %-v: %v", repo, p, err)
return err
}
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
if err := contentStore.Put(p, content); err != nil {
log.Error("Repo[%-v]: Error storing content for LFS meta object %-v: %v", repo, p, err)
if _, err2 := git_model.RemoveLFSMetaObjectByOid(repo.ID, p.Oid); err2 != nil {
log.Error("Repo[%-v]: Error removing LFS meta object %-v: %v", repo, p, err2)
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
}
return err
}
return nil
})
if err != nil {
select {
case <-ctx.Done():
return nil
default:
}
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
}
return err
}
var batch []lfs.Pointer
for pointerBlob := range pointerChan {
meta, err := git_model.GetLFSMetaObjectByOid(repo.ID, pointerBlob.Oid)
if err != nil && err != git_model.ErrLFSObjectNotExist {
log.Error("Repo[%-v]: Error querying LFS meta object %-v: %v", repo, pointerBlob.Pointer, err)
return err
}
if meta != nil {
log.Trace("Repo[%-v]: Skipping unknown LFS meta object %-v", repo, pointerBlob.Pointer)
continue
}
log.Trace("Repo[%-v]: LFS object %-v not present in repository", repo, pointerBlob.Pointer)
exist, err := contentStore.Exists(pointerBlob.Pointer)
if err != nil {
log.Error("Repo[%-v]: Error checking if LFS object %-v exists: %v", repo, pointerBlob.Pointer, err)
return err
}
if exist {
log.Trace("Repo[%-v]: LFS object %-v already present; creating meta object", repo, pointerBlob.Pointer)
_, err := git_model.NewLFSMetaObject(&git_model.LFSMetaObject{Pointer: pointerBlob.Pointer, RepositoryID: repo.ID})
if err != nil {
log.Error("Repo[%-v]: Error creating LFS meta object %-v: %v", repo, pointerBlob.Pointer, err)
return err
}
} else {
if setting.LFS.MaxFileSize > 0 && pointerBlob.Size > setting.LFS.MaxFileSize {
log.Info("Repo[%-v]: LFS object %-v download denied because of LFS_MAX_FILE_SIZE=%d < size %d", repo, pointerBlob.Pointer, setting.LFS.MaxFileSize, pointerBlob.Size)
continue
}
batch = append(batch, pointerBlob.Pointer)
if len(batch) >= lfsClient.BatchSize() {
if err := downloadObjects(batch); err != nil {
return err
}
batch = nil
}
}
}
if len(batch) > 0 {
if err := downloadObjects(batch); err != nil {
return err
}
}
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
err, has := <-errChan
if has {
log.Error("Repo[%-v]: Error enumerating LFS objects for repository: %v", repo, err)
Add LFS Migration and Mirror (#14726) * Implemented LFS client. * Implemented scanning for pointer files. * Implemented downloading of lfs files. * Moved model-dependent code into services. * Removed models dependency. Added TryReadPointerFromBuffer. * Migrated code from service to module. * Centralised storage creation. * Removed dependency from models. * Moved ContentStore into modules. * Share structs between server and client. * Moved method to services. * Implemented lfs download on clone. * Implemented LFS sync on clone and mirror update. * Added form fields. * Updated templates. * Fixed condition. * Use alternate endpoint. * Added missing methods. * Fixed typo and make linter happy. * Detached pointer parser from gogit dependency. * Fixed TestGetLFSRange test. * Added context to support cancellation. * Use ReadFull to probably read more data. * Removed duplicated code from models. * Moved scan implementation into pointer_scanner_nogogit. * Changed method name. * Added comments. * Added more/specific log/error messages. * Embedded lfs.Pointer into models.LFSMetaObject. * Moved code from models to module. * Moved code from models to module. * Moved code from models to module. * Reduced pointer usage. * Embedded type. * Use promoted fields. * Fixed unexpected eof. * Added unit tests. * Implemented migration of local file paths. * Show an error on invalid LFS endpoints. * Hide settings if not used. * Added LFS info to mirror struct. * Fixed comment. * Check LFS endpoint. * Manage LFS settings from mirror page. * Fixed selector. * Adjusted selector. * Added more tests. * Added local filesystem migration test. * Fixed typo. * Reset settings. * Added special windows path handling. * Added unit test for HTTPClient. * Added unit test for BasicTransferAdapter. * Moved into util package. * Test if LFS endpoint is allowed. * Added support for git:// * Just use a static placeholder as the displayed url may be invalid. * Reverted to original code. * Added "Advanced Settings". * Updated wording. * Added discovery info link. * Implemented suggestion. * Fixed missing format parameter. * Added Pointer.IsValid(). * Always remove model on error. * Added suggestions. * Use channel instead of array. * Update routers/repo/migrate.go * fmt Signed-off-by: Andrew Thornton <art27@cantab.net> Co-authored-by: zeripath <art27@cantab.net>
2021-04-09 00:25:57 +02:00
return err
}
return nil
}
Improve sync performance for pull-mirrors (#19125) This addresses https://github.com/go-gitea/gitea/issues/18352 It aims to improve performance (and resource use) of the `SyncReleasesWithTags` operation for pull-mirrors. For large repositories with many tags, `SyncReleasesWithTags` can be a costly operation (taking several minutes to complete). The reason is two-fold: 1. on sync, every upstream repo tag is compared (for changes) against existing local entries in the release table to ensure that they are up-to-date. 2. the procedure for getting _each tag_ involves a series of git operations ```bash git show-ref --tags -- v8.2.4477 git cat-file -t 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git cat-file -p 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git rev-list --count 29ab6ce9f36660cffaad3c8789e71162e5db5d2f ``` of which the `git rev-list --count` can be particularly heavy. This PR optimizes performance for pull-mirrors. We utilize the fact that a pull-mirror is always identical to its upstream and rebuild the entire release table on every sync and use a batch `git for-each-ref .. refs/tags` call to retrieve all tags in one go. For large mirror repos, with hundreds of annotated tags, this brings down the duration of the sync operation from several minutes to a few seconds. A few unscientific examples run on my local machine: - https://github.com/spring-projects/spring-boot (223 tags) - before: `0m28,673s` - after: `0m2,244s` - https://github.com/kubernetes/kubernetes (890 tags) - before: `8m00s` - after: `0m8,520s` - https://github.com/vim/vim (13954 tags) - before: `14m20,383s` - after: `0m35,467s` I added a `foreachref` package which contains a flexible way of specifying which reference fields are of interest (`git-for-each-ref(1)`) and to produce a parser for the expected output. These could be reused in other places where `for-each-ref` is used. I'll add unit tests for those if the overall PR looks promising.
2022-03-31 14:30:40 +02:00
// pullMirrorReleaseSync is a pull-mirror specific tag<->release table
// synchronization which overwrites all Releases from the repository tags. This
// can be relied on since a pull-mirror is always identical to its
// upstream. Hence, after each sync we want the pull-mirror release set to be
// identical to the upstream tag set. This is much more efficient for
// repositories like https://github.com/vim/vim (with over 13000 tags).
func pullMirrorReleaseSync(repo *repo_model.Repository, gitRepo *git.Repository) error {
log.Trace("pullMirrorReleaseSync: rebuilding releases for pull-mirror Repo[%d:%s/%s]", repo.ID, repo.OwnerName, repo.Name)
tags, numTags, err := gitRepo.GetTagInfos(0, 0)
if err != nil {
return fmt.Errorf("unable to GetTagInfos in pull-mirror Repo[%d:%s/%s]: %w", repo.ID, repo.OwnerName, repo.Name, err)
}
err = db.WithTx(func(ctx context.Context) error {
//
// clear out existing releases
//
if _, err := db.DeleteByBean(ctx, &repo_model.Release{RepoID: repo.ID}); err != nil {
Improve sync performance for pull-mirrors (#19125) This addresses https://github.com/go-gitea/gitea/issues/18352 It aims to improve performance (and resource use) of the `SyncReleasesWithTags` operation for pull-mirrors. For large repositories with many tags, `SyncReleasesWithTags` can be a costly operation (taking several minutes to complete). The reason is two-fold: 1. on sync, every upstream repo tag is compared (for changes) against existing local entries in the release table to ensure that they are up-to-date. 2. the procedure for getting _each tag_ involves a series of git operations ```bash git show-ref --tags -- v8.2.4477 git cat-file -t 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git cat-file -p 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git rev-list --count 29ab6ce9f36660cffaad3c8789e71162e5db5d2f ``` of which the `git rev-list --count` can be particularly heavy. This PR optimizes performance for pull-mirrors. We utilize the fact that a pull-mirror is always identical to its upstream and rebuild the entire release table on every sync and use a batch `git for-each-ref .. refs/tags` call to retrieve all tags in one go. For large mirror repos, with hundreds of annotated tags, this brings down the duration of the sync operation from several minutes to a few seconds. A few unscientific examples run on my local machine: - https://github.com/spring-projects/spring-boot (223 tags) - before: `0m28,673s` - after: `0m2,244s` - https://github.com/kubernetes/kubernetes (890 tags) - before: `8m00s` - after: `0m8,520s` - https://github.com/vim/vim (13954 tags) - before: `14m20,383s` - after: `0m35,467s` I added a `foreachref` package which contains a flexible way of specifying which reference fields are of interest (`git-for-each-ref(1)`) and to produce a parser for the expected output. These could be reused in other places where `for-each-ref` is used. I'll add unit tests for those if the overall PR looks promising.
2022-03-31 14:30:40 +02:00
return fmt.Errorf("unable to clear releases for pull-mirror Repo[%d:%s/%s]: %w", repo.ID, repo.OwnerName, repo.Name, err)
}
//
// make release set identical to upstream tags
//
for _, tag := range tags {
release := repo_model.Release{
Improve sync performance for pull-mirrors (#19125) This addresses https://github.com/go-gitea/gitea/issues/18352 It aims to improve performance (and resource use) of the `SyncReleasesWithTags` operation for pull-mirrors. For large repositories with many tags, `SyncReleasesWithTags` can be a costly operation (taking several minutes to complete). The reason is two-fold: 1. on sync, every upstream repo tag is compared (for changes) against existing local entries in the release table to ensure that they are up-to-date. 2. the procedure for getting _each tag_ involves a series of git operations ```bash git show-ref --tags -- v8.2.4477 git cat-file -t 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git cat-file -p 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git rev-list --count 29ab6ce9f36660cffaad3c8789e71162e5db5d2f ``` of which the `git rev-list --count` can be particularly heavy. This PR optimizes performance for pull-mirrors. We utilize the fact that a pull-mirror is always identical to its upstream and rebuild the entire release table on every sync and use a batch `git for-each-ref .. refs/tags` call to retrieve all tags in one go. For large mirror repos, with hundreds of annotated tags, this brings down the duration of the sync operation from several minutes to a few seconds. A few unscientific examples run on my local machine: - https://github.com/spring-projects/spring-boot (223 tags) - before: `0m28,673s` - after: `0m2,244s` - https://github.com/kubernetes/kubernetes (890 tags) - before: `8m00s` - after: `0m8,520s` - https://github.com/vim/vim (13954 tags) - before: `14m20,383s` - after: `0m35,467s` I added a `foreachref` package which contains a flexible way of specifying which reference fields are of interest (`git-for-each-ref(1)`) and to produce a parser for the expected output. These could be reused in other places where `for-each-ref` is used. I'll add unit tests for those if the overall PR looks promising.
2022-03-31 14:30:40 +02:00
RepoID: repo.ID,
TagName: tag.Name,
LowerTagName: strings.ToLower(tag.Name),
Sha1: tag.Object.String(),
// NOTE: ignored, since NumCommits are unused
// for pull-mirrors (only relevant when
// displaying releases, IsTag: false)
NumCommits: -1,
CreatedUnix: timeutil.TimeStamp(tag.Tagger.When.Unix()),
IsTag: true,
}
if err := db.Insert(ctx, release); err != nil {
return fmt.Errorf("unable insert tag %s for pull-mirror Repo[%d:%s/%s]: %w", tag.Name, repo.ID, repo.OwnerName, repo.Name, err)
}
}
return nil
})
if err != nil {
return fmt.Errorf("unable to rebuild release table for pull-mirror Repo[%d:%s/%s]: %w", repo.ID, repo.OwnerName, repo.Name, err)
}
log.Trace("pullMirrorReleaseSync: done rebuilding %d releases", numTags)
return nil
}