1
0
Fork 0
mirror of https://codeberg.org/forgejo/forgejo.git synced 2024-12-30 14:09:42 -05:00
forgejo/services/migrations/github.go

884 lines
24 KiB
Go
Raw Normal View History

// Copyright 2019 The Gitea Authors. All rights reserved.
// Copyright 2018 Jonas Franz. All rights reserved.
// SPDX-License-Identifier: MIT
package migrations
import (
"context"
"fmt"
"io"
"net/http"
"net/url"
"strconv"
"strings"
"time"
"code.gitea.io/gitea/modules/git"
"code.gitea.io/gitea/modules/log"
base "code.gitea.io/gitea/modules/migration"
"code.gitea.io/gitea/modules/proxy"
"code.gitea.io/gitea/modules/structs"
"code.gitea.io/gitea/modules/util"
"github.com/google/go-github/v45/github"
"golang.org/x/oauth2"
)
var (
_ base.Downloader = &GithubDownloaderV3{}
_ base.DownloaderFactory = &GithubDownloaderV3Factory{}
// GithubLimitRateRemaining limit to wait for new rate to apply
GithubLimitRateRemaining = 0
)
func init() {
RegisterDownloaderFactory(&GithubDownloaderV3Factory{})
}
// GithubDownloaderV3Factory defines a github downloader v3 factory
type GithubDownloaderV3Factory struct{}
// New returns a Downloader related to this factory according MigrateOptions
func (f *GithubDownloaderV3Factory) New(ctx context.Context, opts base.MigrateOptions) (base.Downloader, error) {
2019-10-13 09:23:14 -04:00
u, err := url.Parse(opts.CloneAddr)
if err != nil {
return nil, err
}
baseURL := u.Scheme + "://" + u.Host
fields := strings.Split(u.Path, "/")
oldOwner := fields[1]
oldName := strings.TrimSuffix(fields[2], ".git")
log.Trace("Create github downloader BaseURL: %s %s/%s", baseURL, oldOwner, oldName)
return NewGithubDownloaderV3(ctx, baseURL, opts.AuthUsername, opts.AuthPassword, opts.AuthToken, oldOwner, oldName), nil
}
// GitServiceType returns the type of git service
func (f *GithubDownloaderV3Factory) GitServiceType() structs.GitServiceType {
return structs.GithubService
}
// GithubDownloaderV3 implements a Downloader interface to get repository information
// from github via APIv3
type GithubDownloaderV3 struct {
base.NullDownloader
ctx context.Context
clients []*github.Client
baseURL string
repoOwner string
repoName string
userName string
password string
rates []*github.Rate
curClientIdx int
maxPerPage int
SkipReactions bool
SkipReviews bool
}
// NewGithubDownloaderV3 creates a github Downloader via github v3 API
func NewGithubDownloaderV3(ctx context.Context, baseURL, userName, password, token, repoOwner, repoName string) *GithubDownloaderV3 {
downloader := GithubDownloaderV3{
userName: userName,
baseURL: baseURL,
password: password,
ctx: ctx,
repoOwner: repoOwner,
repoName: repoName,
maxPerPage: 100,
}
if token != "" {
tokens := strings.Split(token, ",")
for _, token := range tokens {
token = strings.TrimSpace(token)
ts := oauth2.StaticTokenSource(
&oauth2.Token{AccessToken: token},
)
client := &http.Client{
Transport: &oauth2.Transport{
Base: NewMigrationHTTPTransport(),
Source: oauth2.ReuseTokenSource(nil, ts),
},
}
downloader.addClient(client, baseURL)
}
} else {
transport := NewMigrationHTTPTransport()
transport.Proxy = func(req *http.Request) (*url.URL, error) {
req.SetBasicAuth(userName, password)
return proxy.Proxy()(req)
}
client := &http.Client{
Transport: transport,
}
downloader.addClient(client, baseURL)
}
return &downloader
}
// String implements Stringer
func (g *GithubDownloaderV3) String() string {
return fmt.Sprintf("migration from github server %s %s/%s", g.baseURL, g.repoOwner, g.repoName)
}
// ColorFormat provides a basic color format for a GithubDownloader
func (g *GithubDownloaderV3) ColorFormat(s fmt.State) {
if g == nil {
log.ColorFprintf(s, "<nil: GithubDownloaderV3>")
return
}
log.ColorFprintf(s, "migration from github server %s %s/%s", g.baseURL, g.repoOwner, g.repoName)
}
func (g *GithubDownloaderV3) addClient(client *http.Client, baseURL string) {
githubClient := github.NewClient(client)
if baseURL != "https://github.com" {
githubClient, _ = github.NewEnterpriseClient(baseURL, baseURL, client)
}
g.clients = append(g.clients, githubClient)
g.rates = append(g.rates, nil)
}
// SetContext set context
func (g *GithubDownloaderV3) SetContext(ctx context.Context) {
g.ctx = ctx
}
func (g *GithubDownloaderV3) waitAndPickClient() {
var recentIdx int
var maxRemaining int
for i := 0; i < len(g.clients); i++ {
if g.rates[i] != nil && g.rates[i].Remaining > maxRemaining {
maxRemaining = g.rates[i].Remaining
recentIdx = i
}
}
g.curClientIdx = recentIdx // if no max remain, it will always pick the first client.
for g.rates[g.curClientIdx] != nil && g.rates[g.curClientIdx].Remaining <= GithubLimitRateRemaining {
timer := time.NewTimer(time.Until(g.rates[g.curClientIdx].Reset.Time))
select {
case <-g.ctx.Done():
util.StopTimer(timer)
return
case <-timer.C:
}
err := g.RefreshRate()
if err != nil {
log.Error("g.getClient().RateLimits: %s", err)
}
}
}
// RefreshRate update the current rate (doesn't count in rate limit)
func (g *GithubDownloaderV3) RefreshRate() error {
rates, _, err := g.getClient().RateLimits(g.ctx)
if err != nil {
// if rate limit is not enabled, ignore it
if strings.Contains(err.Error(), "404") {
g.setRate(nil)
return nil
}
return err
}
g.setRate(rates.GetCore())
return nil
}
func (g *GithubDownloaderV3) getClient() *github.Client {
return g.clients[g.curClientIdx]
}
func (g *GithubDownloaderV3) setRate(rate *github.Rate) {
g.rates[g.curClientIdx] = rate
}
// GetRepoInfo returns a repository information
func (g *GithubDownloaderV3) GetRepoInfo() (*base.Repository, error) {
g.waitAndPickClient()
gr, resp, err := g.getClient().Repositories.Get(g.ctx, g.repoOwner, g.repoName)
if err != nil {
return nil, err
}
g.setRate(&resp.Rate)
// convert github repo to stand Repo
return &base.Repository{
Owner: g.repoOwner,
Name: gr.GetName(),
IsPrivate: gr.GetPrivate(),
Description: gr.GetDescription(),
OriginalURL: gr.GetHTMLURL(),
CloneURL: gr.GetCloneURL(),
DefaultBranch: gr.GetDefaultBranch(),
}, nil
}
// GetTopics return github topics
func (g *GithubDownloaderV3) GetTopics() ([]string, error) {
g.waitAndPickClient()
r, resp, err := g.getClient().Repositories.Get(g.ctx, g.repoOwner, g.repoName)
if err != nil {
return nil, err
}
g.setRate(&resp.Rate)
return r.Topics, nil
}
// GetMilestones returns milestones
func (g *GithubDownloaderV3) GetMilestones() ([]*base.Milestone, error) {
perPage := g.maxPerPage
milestones := make([]*base.Milestone, 0, perPage)
for i := 1; ; i++ {
g.waitAndPickClient()
ms, resp, err := g.getClient().Issues.ListMilestones(g.ctx, g.repoOwner, g.repoName,
&github.MilestoneListOptions{
State: "all",
ListOptions: github.ListOptions{
Page: i,
PerPage: perPage,
},
})
if err != nil {
return nil, err
}
g.setRate(&resp.Rate)
for _, m := range ms {
state := "open"
if m.State != nil {
state = *m.State
}
milestones = append(milestones, &base.Milestone{
Title: m.GetTitle(),
Description: m.GetDescription(),
Deadline: m.DueOn,
State: state,
Created: m.GetCreatedAt(),
Updated: m.UpdatedAt,
Closed: m.ClosedAt,
})
}
if len(ms) < perPage {
break
}
}
return milestones, nil
}
func convertGithubLabel(label *github.Label) *base.Label {
return &base.Label{
Name: label.GetName(),
Color: label.GetColor(),
Description: label.GetDescription(),
}
}
// GetLabels returns labels
func (g *GithubDownloaderV3) GetLabels() ([]*base.Label, error) {
perPage := g.maxPerPage
labels := make([]*base.Label, 0, perPage)
for i := 1; ; i++ {
g.waitAndPickClient()
ls, resp, err := g.getClient().Issues.ListLabels(g.ctx, g.repoOwner, g.repoName,
&github.ListOptions{
Page: i,
PerPage: perPage,
})
if err != nil {
return nil, err
}
g.setRate(&resp.Rate)
for _, label := range ls {
labels = append(labels, convertGithubLabel(label))
}
if len(ls) < perPage {
break
}
}
return labels, nil
}
func (g *GithubDownloaderV3) convertGithubRelease(rel *github.RepositoryRelease) *base.Release {
// GitHub allows commitish to be a reference.
// In this case, we need to remove the prefix, i.e. convert "refs/heads/main" to "main".
targetCommitish := strings.TrimPrefix(rel.GetTargetCommitish(), git.BranchPrefix)
r := &base.Release{
Name: rel.GetName(),
TagName: rel.GetTagName(),
TargetCommitish: targetCommitish,
Draft: rel.GetDraft(),
Prerelease: rel.GetPrerelease(),
Created: rel.GetCreatedAt().Time,
PublisherID: rel.GetAuthor().GetID(),
PublisherName: rel.GetAuthor().GetLogin(),
PublisherEmail: rel.GetAuthor().GetEmail(),
Body: rel.GetBody(),
}
if rel.PublishedAt != nil {
r.Published = rel.PublishedAt.Time
}
httpClient := NewMigrationHTTPClient()
for _, asset := range rel.Assets {
assetID := *asset.ID // Don't optimize this, for closure we need a local variable
r.Assets = append(r.Assets, &base.ReleaseAsset{
ID: asset.GetID(),
Name: asset.GetName(),
ContentType: asset.ContentType,
Size: asset.Size,
DownloadCount: asset.DownloadCount,
Created: asset.CreatedAt.Time,
Updated: asset.UpdatedAt.Time,
DownloadFunc: func() (io.ReadCloser, error) {
g.waitAndPickClient()
readCloser, redirectURL, err := g.getClient().Repositories.DownloadReleaseAsset(g.ctx, g.repoOwner, g.repoName, assetID, nil)
if err != nil {
return nil, err
}
if err := g.RefreshRate(); err != nil {
log.Error("g.getClient().RateLimits: %s", err)
}
if readCloser != nil {
return readCloser, nil
}
if redirectURL == "" {
return nil, fmt.Errorf("no release asset found for %d", assetID)
}
// Prevent open redirect
if !hasBaseURL(redirectURL, g.baseURL) &&
!hasBaseURL(redirectURL, "https://objects.githubusercontent.com/") {
WarnAndNotice("Unexpected AssetURL for assetID[%d] in %s: %s", asset.GetID(), g, redirectURL)
return io.NopCloser(strings.NewReader(redirectURL)), nil
}
g.waitAndPickClient()
req, err := http.NewRequestWithContext(g.ctx, "GET", redirectURL, nil)
if err != nil {
return nil, err
}
resp, err := httpClient.Do(req)
err1 := g.RefreshRate()
if err1 != nil {
log.Error("g.RefreshRate(): %s", err1)
}
if err != nil {
return nil, err
}
return resp.Body, nil
},
})
}
return r
}
// GetReleases returns releases
func (g *GithubDownloaderV3) GetReleases() ([]*base.Release, error) {
perPage := g.maxPerPage
releases := make([]*base.Release, 0, perPage)
for i := 1; ; i++ {
g.waitAndPickClient()
ls, resp, err := g.getClient().Repositories.ListReleases(g.ctx, g.repoOwner, g.repoName,
&github.ListOptions{
Page: i,
PerPage: perPage,
})
if err != nil {
return nil, err
}
g.setRate(&resp.Rate)
for _, release := range ls {
releases = append(releases, g.convertGithubRelease(release))
}
if len(ls) < perPage {
break
}
}
return releases, nil
}
// GetIssues returns issues according start and limit
func (g *GithubDownloaderV3) GetIssues(page, perPage int) ([]*base.Issue, bool, error) {
if perPage > g.maxPerPage {
perPage = g.maxPerPage
}
opt := &github.IssueListByRepoOptions{
Sort: "created",
Direction: "asc",
State: "all",
ListOptions: github.ListOptions{
PerPage: perPage,
Page: page,
},
}
allIssues := make([]*base.Issue, 0, perPage)
g.waitAndPickClient()
issues, resp, err := g.getClient().Issues.ListByRepo(g.ctx, g.repoOwner, g.repoName, opt)
if err != nil {
return nil, false, fmt.Errorf("error while listing repos: %w", err)
}
log.Trace("Request get issues %d/%d, but in fact get %d", perPage, page, len(issues))
g.setRate(&resp.Rate)
for _, issue := range issues {
if issue.IsPullRequest() {
continue
}
labels := make([]*base.Label, 0, len(issue.Labels))
for _, l := range issue.Labels {
labels = append(labels, convertGithubLabel(l))
}
// get reactions
var reactions []*base.Reaction
if !g.SkipReactions {
for i := 1; ; i++ {
g.waitAndPickClient()
res, resp, err := g.getClient().Reactions.ListIssueReactions(g.ctx, g.repoOwner, g.repoName, issue.GetNumber(), &github.ListOptions{
Page: i,
PerPage: perPage,
})
if err != nil {
return nil, false, err
}
g.setRate(&resp.Rate)
if len(res) == 0 {
break
}
for _, reaction := range res {
reactions = append(reactions, &base.Reaction{
UserID: reaction.User.GetID(),
UserName: reaction.User.GetLogin(),
Content: reaction.GetContent(),
})
}
}
}
var assignees []string
for i := range issue.Assignees {
assignees = append(assignees, issue.Assignees[i].GetLogin())
}
allIssues = append(allIssues, &base.Issue{
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
Title: *issue.Title,
Number: int64(*issue.Number),
PosterID: issue.GetUser().GetID(),
PosterName: issue.GetUser().GetLogin(),
PosterEmail: issue.GetUser().GetEmail(),
Content: issue.GetBody(),
Milestone: issue.GetMilestone().GetTitle(),
State: issue.GetState(),
Created: issue.GetCreatedAt(),
Updated: issue.GetUpdatedAt(),
Labels: labels,
Reactions: reactions,
Closed: issue.ClosedAt,
IsLocked: issue.GetLocked(),
Assignees: assignees,
ForeignIndex: int64(*issue.Number),
})
}
return allIssues, len(issues) < perPage, nil
}
// SupportGetRepoComments return true if it supports get repo comments
func (g *GithubDownloaderV3) SupportGetRepoComments() bool {
return true
}
// GetComments returns comments according issueNumber
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
func (g *GithubDownloaderV3) GetComments(commentable base.Commentable) ([]*base.Comment, bool, error) {
comments, err := g.getComments(commentable)
return comments, false, err
}
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
func (g *GithubDownloaderV3) getComments(commentable base.Commentable) ([]*base.Comment, error) {
var (
allComments = make([]*base.Comment, 0, g.maxPerPage)
created = "created"
asc = "asc"
)
opt := &github.IssueListCommentsOptions{
Sort: &created,
Direction: &asc,
ListOptions: github.ListOptions{
PerPage: g.maxPerPage,
},
}
for {
g.waitAndPickClient()
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
comments, resp, err := g.getClient().Issues.ListComments(g.ctx, g.repoOwner, g.repoName, int(commentable.GetForeignIndex()), opt)
if err != nil {
return nil, fmt.Errorf("error while listing repos: %w", err)
}
g.setRate(&resp.Rate)
for _, comment := range comments {
// get reactions
var reactions []*base.Reaction
if !g.SkipReactions {
for i := 1; ; i++ {
g.waitAndPickClient()
res, resp, err := g.getClient().Reactions.ListIssueCommentReactions(g.ctx, g.repoOwner, g.repoName, comment.GetID(), &github.ListOptions{
Page: i,
PerPage: g.maxPerPage,
})
if err != nil {
return nil, err
}
g.setRate(&resp.Rate)
if len(res) == 0 {
break
}
for _, reaction := range res {
reactions = append(reactions, &base.Reaction{
UserID: reaction.User.GetID(),
UserName: reaction.User.GetLogin(),
Content: reaction.GetContent(),
})
}
}
}
allComments = append(allComments, &base.Comment{
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
IssueIndex: commentable.GetLocalIndex(),
Index: comment.GetID(),
PosterID: comment.GetUser().GetID(),
PosterName: comment.GetUser().GetLogin(),
PosterEmail: comment.GetUser().GetEmail(),
Content: comment.GetBody(),
Created: comment.GetCreatedAt(),
Updated: comment.GetUpdatedAt(),
Reactions: reactions,
})
}
if resp.NextPage == 0 {
break
}
opt.Page = resp.NextPage
}
return allComments, nil
}
// GetAllComments returns repository comments according page and perPageSize
func (g *GithubDownloaderV3) GetAllComments(page, perPage int) ([]*base.Comment, bool, error) {
var (
allComments = make([]*base.Comment, 0, perPage)
created = "created"
asc = "asc"
)
if perPage > g.maxPerPage {
perPage = g.maxPerPage
}
opt := &github.IssueListCommentsOptions{
Sort: &created,
Direction: &asc,
ListOptions: github.ListOptions{
Page: page,
PerPage: perPage,
},
}
g.waitAndPickClient()
comments, resp, err := g.getClient().Issues.ListComments(g.ctx, g.repoOwner, g.repoName, 0, opt)
if err != nil {
return nil, false, fmt.Errorf("error while listing repos: %w", err)
}
isEnd := resp.NextPage == 0
log.Trace("Request get comments %d/%d, but in fact get %d, next page is %d", perPage, page, len(comments), resp.NextPage)
g.setRate(&resp.Rate)
for _, comment := range comments {
// get reactions
var reactions []*base.Reaction
if !g.SkipReactions {
for i := 1; ; i++ {
g.waitAndPickClient()
res, resp, err := g.getClient().Reactions.ListIssueCommentReactions(g.ctx, g.repoOwner, g.repoName, comment.GetID(), &github.ListOptions{
Page: i,
PerPage: g.maxPerPage,
})
if err != nil {
return nil, false, err
}
g.setRate(&resp.Rate)
if len(res) == 0 {
break
}
for _, reaction := range res {
reactions = append(reactions, &base.Reaction{
UserID: reaction.User.GetID(),
UserName: reaction.User.GetLogin(),
Content: reaction.GetContent(),
})
}
}
}
idx := strings.LastIndex(*comment.IssueURL, "/")
issueIndex, _ := strconv.ParseInt((*comment.IssueURL)[idx+1:], 10, 64)
allComments = append(allComments, &base.Comment{
IssueIndex: issueIndex,
Index: comment.GetID(),
PosterID: comment.GetUser().GetID(),
PosterName: comment.GetUser().GetLogin(),
PosterEmail: comment.GetUser().GetEmail(),
Content: comment.GetBody(),
Created: comment.GetCreatedAt(),
Updated: comment.GetUpdatedAt(),
Reactions: reactions,
})
}
return allComments, isEnd, nil
}
// GetPullRequests returns pull requests according page and perPage
func (g *GithubDownloaderV3) GetPullRequests(page, perPage int) ([]*base.PullRequest, bool, error) {
if perPage > g.maxPerPage {
perPage = g.maxPerPage
}
opt := &github.PullRequestListOptions{
Sort: "created",
Direction: "asc",
State: "all",
ListOptions: github.ListOptions{
PerPage: perPage,
Page: page,
},
}
allPRs := make([]*base.PullRequest, 0, perPage)
g.waitAndPickClient()
prs, resp, err := g.getClient().PullRequests.List(g.ctx, g.repoOwner, g.repoName, opt)
if err != nil {
return nil, false, fmt.Errorf("error while listing repos: %w", err)
}
log.Trace("Request get pull requests %d/%d, but in fact get %d", perPage, page, len(prs))
g.setRate(&resp.Rate)
for _, pr := range prs {
labels := make([]*base.Label, 0, len(pr.Labels))
for _, l := range pr.Labels {
labels = append(labels, convertGithubLabel(l))
}
// get reactions
var reactions []*base.Reaction
if !g.SkipReactions {
for i := 1; ; i++ {
g.waitAndPickClient()
res, resp, err := g.getClient().Reactions.ListIssueReactions(g.ctx, g.repoOwner, g.repoName, pr.GetNumber(), &github.ListOptions{
Page: i,
PerPage: perPage,
})
if err != nil {
return nil, false, err
}
g.setRate(&resp.Rate)
if len(res) == 0 {
break
}
for _, reaction := range res {
reactions = append(reactions, &base.Reaction{
UserID: reaction.User.GetID(),
UserName: reaction.User.GetLogin(),
Content: reaction.GetContent(),
})
}
}
}
// download patch and saved as tmp file
g.waitAndPickClient()
allPRs = append(allPRs, &base.PullRequest{
Title: pr.GetTitle(),
Number: int64(pr.GetNumber()),
PosterID: pr.GetUser().GetID(),
PosterName: pr.GetUser().GetLogin(),
PosterEmail: pr.GetUser().GetEmail(),
Content: pr.GetBody(),
Milestone: pr.GetMilestone().GetTitle(),
State: pr.GetState(),
Created: pr.GetCreatedAt(),
Updated: pr.GetUpdatedAt(),
Closed: pr.ClosedAt,
Labels: labels,
Merged: pr.MergedAt != nil,
MergeCommitSHA: pr.GetMergeCommitSHA(),
MergedTime: pr.MergedAt,
IsLocked: pr.ActiveLockReason != nil,
Head: base.PullRequestBranch{
Ref: pr.GetHead().GetRef(),
SHA: pr.GetHead().GetSHA(),
OwnerName: pr.GetHead().GetUser().GetLogin(),
RepoName: pr.GetHead().GetRepo().GetName(),
CloneURL: pr.GetHead().GetRepo().GetCloneURL(), // see below for SECURITY related issues here
},
Base: base.PullRequestBranch{
Ref: pr.GetBase().GetRef(),
SHA: pr.GetBase().GetSHA(),
RepoName: pr.GetBase().GetRepo().GetName(),
OwnerName: pr.GetBase().GetUser().GetLogin(),
},
PatchURL: pr.GetPatchURL(), // see below for SECURITY related issues here
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
Reactions: reactions,
ForeignIndex: int64(*pr.Number),
})
// SECURITY: Ensure that the PR is safe
_ = CheckAndEnsureSafePR(allPRs[len(allPRs)-1], g.baseURL, g)
}
return allPRs, len(prs) < perPage, nil
}
func convertGithubReview(r *github.PullRequestReview) *base.Review {
return &base.Review{
ID: r.GetID(),
ReviewerID: r.GetUser().GetID(),
ReviewerName: r.GetUser().GetLogin(),
CommitID: r.GetCommitID(),
Content: r.GetBody(),
CreatedAt: r.GetSubmittedAt(),
State: r.GetState(),
}
}
func (g *GithubDownloaderV3) convertGithubReviewComments(cs []*github.PullRequestComment) ([]*base.ReviewComment, error) {
rcs := make([]*base.ReviewComment, 0, len(cs))
for _, c := range cs {
// get reactions
var reactions []*base.Reaction
if !g.SkipReactions {
for i := 1; ; i++ {
g.waitAndPickClient()
res, resp, err := g.getClient().Reactions.ListPullRequestCommentReactions(g.ctx, g.repoOwner, g.repoName, c.GetID(), &github.ListOptions{
Page: i,
PerPage: g.maxPerPage,
})
if err != nil {
return nil, err
}
g.setRate(&resp.Rate)
if len(res) == 0 {
break
}
for _, reaction := range res {
reactions = append(reactions, &base.Reaction{
UserID: reaction.User.GetID(),
UserName: reaction.User.GetLogin(),
Content: reaction.GetContent(),
})
}
}
}
rcs = append(rcs, &base.ReviewComment{
ID: c.GetID(),
InReplyTo: c.GetInReplyTo(),
Content: c.GetBody(),
TreePath: c.GetPath(),
DiffHunk: c.GetDiffHunk(),
Position: c.GetPosition(),
CommitID: c.GetCommitID(),
PosterID: c.GetUser().GetID(),
Reactions: reactions,
CreatedAt: c.GetCreatedAt(),
UpdatedAt: c.GetUpdatedAt(),
})
}
return rcs, nil
}
// GetReviews returns pull requests review
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
func (g *GithubDownloaderV3) GetReviews(reviewable base.Reviewable) ([]*base.Review, error) {
allReviews := make([]*base.Review, 0, g.maxPerPage)
if g.SkipReviews {
return allReviews, nil
}
opt := &github.ListOptions{
PerPage: g.maxPerPage,
}
// Get approve/request change reviews
for {
g.waitAndPickClient()
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
reviews, resp, err := g.getClient().PullRequests.ListReviews(g.ctx, g.repoOwner, g.repoName, int(reviewable.GetForeignIndex()), opt)
if err != nil {
return nil, fmt.Errorf("error while listing repos: %w", err)
}
g.setRate(&resp.Rate)
for _, review := range reviews {
r := convertGithubReview(review)
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
r.IssueIndex = reviewable.GetLocalIndex()
// retrieve all review comments
opt2 := &github.ListOptions{
PerPage: g.maxPerPage,
}
for {
g.waitAndPickClient()
Store the foreign ID of issues during migration (#18446) Storing the foreign identifier of an imported issue in the database is a prerequisite to implement idempotent migrations or mirror for issues. It is a baby step towards mirroring that introduces a new table. At the moment when an issue is created by the Gitea uploader, it fails if the issue already exists. The Gitea uploader could be modified so that, instead of failing, it looks up the database to find an existing issue. And if it does it would update the issue instead of creating a new one. However this is not currently possible because an information is missing from the database: the foreign identifier that uniquely represents the issue being migrated is not persisted. With this change, the foreign identifier is stored in the database and the Gitea uploader will then be able to run a query to figure out if a given issue being imported already exists. The implementation of mirroring for issues, pull requests, releases, etc. can be done in three steps: 1. Store an identifier for the element being mirrored (issue, pull request...) in the database (this is the purpose of these changes) 2. Modify the Gitea uploader to be able to update an existing repository with all it contains (issues, pull request...) instead of failing if it exists 3. Optimize the Gitea uploader to speed up the updates, when possible. The second step creates code that does not yet exist to enable idempotent migrations with the Gitea uploader. When a migration is done for the first time, the behavior is not changed. But when a migration is done for a repository that already exists, this new code is used to update it. The third step can use the code created in the second step to optimize and speed up migrations. For instance, when a migration is resumed, an issue that has an update time that is not more recent can be skipped and only newly created issues or updated ones will be updated. Another example of optimization could be that a webhook notifies Gitea when an issue is updated. The code triggered by the webhook would download only this issue and call the code created in the second step to update the issue, as if it was in the process of an idempotent migration. The ForeignReferences table is added to contain local and foreign ID pairs relative to a given repository. It can later be used for pull requests and other artifacts that can be mirrored. Although the foreign id could be added as a single field in issues or pull requests, it would need to be added to all tables that represent something that can be mirrored. Creating a new table makes for a simpler and more generic design. The drawback is that it requires an extra lookup to obtain the information. However, this extra information is only required during migration or mirroring and does not impact the way Gitea currently works. The foreign identifier of an issue or pull request is similar to the identifier of an external user, which is stored in reactions, issues, etc. as OriginalPosterID and so on. The representation of a user is however different and the ability of users to link their account to an external user at a later time is also a logic that is different from what is involved in mirroring or migrations. For these reasons, despite some commonalities, it is unclear at this time how the two tables (foreign reference and external user) could be merged together. The ForeignID field is extracted from the issue migration context so that it can be dumped in files with dump-repo and later restored via restore-repo. The GetAllComments downloader method is introduced to simplify the implementation and not overload the Context for the purpose of pagination. It also clarifies in which context the comments are paginated and in which context they are not. The Context interface is no longer useful for the purpose of retrieving the LocalID and ForeignID since they are now both available from the PullRequest and Issue struct. The Reviewable and Commentable interfaces replace and serve the same purpose. The Context data member of PullRequest and Issue becomes a DownloaderContext to clarify that its purpose is not to support in memory operations while the current downloader is acting but is not otherwise persisted. It is, for instance, used by the GitLab downloader to store the IsMergeRequest boolean and sort out issues. --- [source](https://lab.forgefriends.org/forgefriends/forgefriends/-/merge_requests/36) Signed-off-by: Loïc Dachary <loic@dachary.org> Co-authored-by: Loïc Dachary <loic@dachary.org>
2022-03-17 13:08:35 -04:00
reviewComments, resp, err := g.getClient().PullRequests.ListReviewComments(g.ctx, g.repoOwner, g.repoName, int(reviewable.GetForeignIndex()), review.GetID(), opt2)
if err != nil {
return nil, fmt.Errorf("error while listing repos: %w", err)
}
g.setRate(&resp.Rate)
cs, err := g.convertGithubReviewComments(reviewComments)
if err != nil {
return nil, err
}
r.Comments = append(r.Comments, cs...)
if resp.NextPage == 0 {
break
}
opt2.Page = resp.NextPage
}
allReviews = append(allReviews, r)
}
if resp.NextPage == 0 {
break
}
opt.Page = resp.NextPage
}
// Get requested reviews
for {
g.waitAndPickClient()
reviewers, resp, err := g.getClient().PullRequests.ListReviewers(g.ctx, g.repoOwner, g.repoName, int(reviewable.GetForeignIndex()), opt)
if err != nil {
return nil, fmt.Errorf("error while listing repos: %w", err)
}
g.setRate(&resp.Rate)
for _, user := range reviewers.Users {
r := &base.Review{
ReviewerID: user.GetID(),
ReviewerName: user.GetLogin(),
State: base.ReviewStateRequestReview,
IssueIndex: reviewable.GetLocalIndex(),
}
allReviews = append(allReviews, r)
}
// TODO: Handle Team requests
if resp.NextPage == 0 {
break
}
opt.Page = resp.NextPage
}
return allReviews, nil
}