1
0
Fork 0
mirror of https://codeberg.org/forgejo/forgejo.git synced 2024-12-14 11:48:09 -05:00
forgejo/services/webhook/deliver_test.go

333 lines
8.7 KiB
Go
Raw Normal View History

// Copyright 2019 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package webhook
import (
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
"context"
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
"io"
"net/http"
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
"net/http/httptest"
"net/url"
"os"
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
"strings"
"testing"
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
"time"
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
"code.gitea.io/gitea/models/db"
"code.gitea.io/gitea/models/unittest"
webhook_model "code.gitea.io/gitea/models/webhook"
"code.gitea.io/gitea/modules/hostmatcher"
"code.gitea.io/gitea/modules/setting"
webhook_module "code.gitea.io/gitea/modules/webhook"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestWebhookProxy(t *testing.T) {
oldWebhook := setting.Webhook
oldHTTPProxy := os.Getenv("http_proxy")
oldHTTPSProxy := os.Getenv("https_proxy")
t.Cleanup(func() {
setting.Webhook = oldWebhook
os.Setenv("http_proxy", oldHTTPProxy)
os.Setenv("https_proxy", oldHTTPSProxy)
})
os.Unsetenv("http_proxy")
os.Unsetenv("https_proxy")
setting.Webhook.ProxyURL = "http://localhost:8080"
setting.Webhook.ProxyURLFixed, _ = url.Parse(setting.Webhook.ProxyURL)
setting.Webhook.ProxyHosts = []string{"*.discordapp.com", "discordapp.com"}
allowedHostMatcher := hostmatcher.ParseHostMatchList("webhook.ALLOWED_HOST_LIST", "discordapp.com,s.discordapp.com")
tests := []struct {
req string
want string
wantErr bool
}{
{
req: "https://discordapp.com/api/webhooks/xxxxxxxxx/xxxxxxxxxxxxxxxxxxx",
want: "http://localhost:8080",
wantErr: false,
},
{
req: "http://s.discordapp.com/assets/xxxxxx",
want: "http://localhost:8080",
wantErr: false,
},
{
req: "http://github.com/a/b",
want: "",
wantErr: false,
},
{
req: "http://www.discordapp.com/assets/xxxxxx",
want: "",
wantErr: true,
},
}
for _, tt := range tests {
t.Run(tt.req, func(t *testing.T) {
req, err := http.NewRequest("POST", tt.req, nil)
require.NoError(t, err)
u, err := webhookProxy(allowedHostMatcher)(req)
if tt.wantErr {
require.Error(t, err)
return
}
require.NoError(t, err)
got := ""
if u != nil {
got = u.String()
}
assert.Equal(t, tt.want, got)
})
}
}
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
func TestWebhookDeliverAuthorizationHeader(t *testing.T) {
require.NoError(t, unittest.PrepareTestDatabase())
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
done := make(chan struct{}, 1)
s := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
assert.Equal(t, "/webhook", r.URL.Path)
assert.Equal(t, "Bearer s3cr3t-t0ken", r.Header.Get("Authorization"))
w.WriteHeader(200)
done <- struct{}{}
}))
t.Cleanup(s.Close)
hook := &webhook_model.Webhook{
RepoID: 3,
URL: s.URL + "/webhook",
ContentType: webhook_model.ContentTypeJSON,
IsActive: true,
Type: webhook_module.GITEA,
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
}
err := hook.SetHeaderAuthorization("Bearer s3cr3t-t0ken")
require.NoError(t, err)
require.NoError(t, webhook_model.CreateWebhook(db.DefaultContext, hook))
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
hookTask := &webhook_model.HookTask{
HookID: hook.ID,
EventType: webhook_module.HookEventPush,
PayloadVersion: 2,
}
hookTask, err = webhook_model.CreateHookTask(db.DefaultContext, hookTask)
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
assert.NotNil(t, hookTask)
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
require.NoError(t, Deliver(context.Background(), hookTask))
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
select {
case <-done:
case <-time.After(5 * time.Second):
t.Fatal("waited to long for request to happen")
}
assert.True(t, hookTask.IsSucceed)
assert.Equal(t, "Bearer ******", hookTask.RequestInfo.Headers["Authorization"])
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
}
func TestWebhookDeliverHookTask(t *testing.T) {
require.NoError(t, unittest.PrepareTestDatabase())
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
done := make(chan struct{}, 1)
s := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
assert.Equal(t, "PUT", r.Method)
switch r.URL.Path {
case "/webhook/66d222a5d6349e1311f551e50722d837e30fce98":
// Version 1
assert.Equal(t, "push", r.Header.Get("X-GitHub-Event"))
assert.Equal(t, "", r.Header.Get("Content-Type"))
body, err := io.ReadAll(r.Body)
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
assert.Equal(t, `{"data": 42}`, string(body))
case "/webhook/6db5dc1e282529a8c162c7fe93dd2667494eeb51":
// Version 2
assert.Equal(t, "application/json", r.Header.Get("Content-Type"))
body, err := io.ReadAll(r.Body)
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
assert.Len(t, body, 2147)
default:
w.WriteHeader(404)
t.Fatalf("unexpected url path %s", r.URL.Path)
return
}
w.WriteHeader(200)
done <- struct{}{}
}))
t.Cleanup(s.Close)
hook := &webhook_model.Webhook{
RepoID: 3,
IsActive: true,
Type: webhook_module.MATRIX,
URL: s.URL + "/webhook",
HTTPMethod: "PUT",
ContentType: webhook_model.ContentTypeJSON,
Meta: `{"message_type":0}`, // text
}
require.NoError(t, webhook_model.CreateWebhook(db.DefaultContext, hook))
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
t.Run("Version 1", func(t *testing.T) {
hookTask := &webhook_model.HookTask{
HookID: hook.ID,
EventType: webhook_module.HookEventPush,
PayloadContent: `{"data": 42}`,
PayloadVersion: 1,
}
hookTask, err := webhook_model.CreateHookTask(db.DefaultContext, hookTask)
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
assert.NotNil(t, hookTask)
require.NoError(t, Deliver(context.Background(), hookTask))
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
select {
case <-done:
case <-time.After(5 * time.Second):
t.Fatal("waited to long for request to happen")
}
assert.True(t, hookTask.IsSucceed)
})
t.Run("Version 2", func(t *testing.T) {
p := pushTestPayload()
data, err := p.JSONPayload()
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
hookTask := &webhook_model.HookTask{
HookID: hook.ID,
EventType: webhook_module.HookEventPush,
PayloadContent: string(data),
PayloadVersion: 2,
}
hookTask, err = webhook_model.CreateHookTask(db.DefaultContext, hookTask)
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
assert.NotNil(t, hookTask)
require.NoError(t, Deliver(context.Background(), hookTask))
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
select {
case <-done:
case <-time.After(5 * time.Second):
t.Fatal("waited to long for request to happen")
}
assert.True(t, hookTask.IsSucceed)
})
}
func TestWebhookDeliverSpecificTypes(t *testing.T) {
require.NoError(t, unittest.PrepareTestDatabase())
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
type hookCase struct {
gotBody chan []byte
expectedMethod string
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
}
cases := map[string]hookCase{
webhook_module.SLACK: {
gotBody: make(chan []byte, 1),
},
webhook_module.DISCORD: {
gotBody: make(chan []byte, 1),
},
webhook_module.DINGTALK: {
gotBody: make(chan []byte, 1),
},
webhook_module.TELEGRAM: {
gotBody: make(chan []byte, 1),
},
webhook_module.MSTEAMS: {
gotBody: make(chan []byte, 1),
},
webhook_module.FEISHU: {
gotBody: make(chan []byte, 1),
},
webhook_module.MATRIX: {
gotBody: make(chan []byte, 1),
expectedMethod: "PUT",
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
},
webhook_module.WECHATWORK: {
gotBody: make(chan []byte, 1),
},
webhook_module.PACKAGIST: {
gotBody: make(chan []byte, 1),
},
}
s := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
assert.Equal(t, "application/json", r.Header.Get("Content-Type"), r.URL.Path)
typ := strings.Split(r.URL.Path, "/")[1] // take first segment (after skipping leading slash)
hc := cases[typ]
if hc.expectedMethod != "" {
assert.Equal(t, hc.expectedMethod, r.Method, r.URL.Path)
} else {
assert.Equal(t, "POST", r.Method, r.URL.Path)
}
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
require.NotNil(t, hc.gotBody, r.URL.Path)
body, err := io.ReadAll(r.Body)
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
w.WriteHeader(200)
hc.gotBody <- body
}))
t.Cleanup(s.Close)
p := pushTestPayload()
data, err := p.JSONPayload()
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
for typ, hc := range cases {
typ := typ
hc := hc
t.Run(typ, func(t *testing.T) {
t.Parallel()
hook := &webhook_model.Webhook{
RepoID: 3,
IsActive: true,
Type: typ,
URL: s.URL + "/" + typ,
HTTPMethod: "", // should fallback to POST, when left unset by the specific hook
ContentType: 0, // set to 0 so that falling back to default request fails with "invalid content type"
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
Meta: "{}",
}
require.NoError(t, webhook_model.CreateWebhook(db.DefaultContext, hook))
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
hookTask := &webhook_model.HookTask{
HookID: hook.ID,
EventType: webhook_module.HookEventPush,
PayloadContent: string(data),
PayloadVersion: 2,
}
hookTask, err := webhook_model.CreateHookTask(db.DefaultContext, hookTask)
require.NoError(t, err)
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
assert.NotNil(t, hookTask)
require.NoError(t, Deliver(context.Background(), hookTask))
Store webhook event in database (#29145) Refactor the webhook logic, to have the type-dependent processing happen only in one place. --- 1. An event happens 2. It is pre-processed (depending on the webhook type) and its body is added to a task queue 3. When the task is processed, some more logic (depending on the webhook type as well) is applied to make an HTTP request This means that webhook-type dependant logic is needed in step 2 and 3. This is cumbersome and brittle to maintain. Updated webhook flow with this PR: 1. An event happens 2. It is stored as-is and added to a task queue 3. When the task is processed, the event is processed (depending on the webhook type) to make an HTTP request So the only webhook-type dependent logic happens in one place (step 3) which should be much more robust. - the raw event must be stored in the hooktask (until now, the pre-processed body was stored) - to ensure that previous hooktasks are correctly sent, a `payload_version` is added (version 1: the body has already been pre-process / version 2: the body is the raw event) So future webhook additions will only have to deal with creating an http.Request based on the raw event (no need to adjust the code in multiple places, like currently). Moreover since this processing happens when fetching from the task queue, it ensures that the queuing of new events (upon a `git push` for instance) does not get slowed down by a slow webhook. As a concrete example, the PR #19307 for custom webhooks, should be substantially smaller: - no need to change `services/webhook/deliver.go` - minimal change in `services/webhook/webhook.go` (add the new webhook to the map) - no need to change all the individual webhook files (since with this refactor the `*webhook_model.Webhook` is provided as argument) (cherry picked from commit 26653b196bd1d15c532af41f60351596dd4330bd) Conflicts: services/webhook/deliver_test.go trivial context conflict
2024-03-07 17:18:38 -05:00
select {
case gotBody := <-hc.gotBody:
assert.NotEqual(t, string(data), string(gotBody), "request body must be different from the event payload")
assert.Equal(t, hookTask.RequestInfo.Body, string(gotBody), "request body was not saved")
case <-time.After(5 * time.Second):
t.Fatal("waited to long for request to happen")
}
assert.True(t, hookTask.IsSucceed)
})
}
Add Webhook authorization header (#20926) _This is a different approach to #20267, I took the liberty of adapting some parts, see below_ ## Context In some cases, a weebhook endpoint requires some kind of authentication. The usual way is by sending a static `Authorization` header, with a given token. For instance: - Matrix expects a `Bearer <token>` (already implemented, by storing the header cleartext in the metadata - which is buggy on retry #19872) - TeamCity #18667 - Gitea instances #20267 - SourceHut https://man.sr.ht/graphql.md#authentication-strategies (this is my actual personal need :) ## Proposed solution Add a dedicated encrypt column to the webhook table (instead of storing it as meta as proposed in #20267), so that it gets available for all present and future hook types (especially the custom ones #19307). This would also solve the buggy matrix retry #19872. As a first step, I would recommend focusing on the backend logic and improve the frontend at a later stage. For now the UI is a simple `Authorization` field (which could be later customized with `Bearer` and `Basic` switches): ![2022-08-23-142911](https://user-images.githubusercontent.com/3864879/186162483-5b721504-eef5-4932-812e-eb96a68494cc.png) The header name is hard-coded, since I couldn't fine any usecase justifying otherwise. ## Questions - What do you think of this approach? @justusbunsi @Gusted @silverwind - ~~How are the migrations generated? Do I have to manually create a new file, or is there a command for that?~~ - ~~I started adding it to the API: should I complete it or should I drop it? (I don't know how much the API is actually used)~~ ## Done as well: - add a migration for the existing matrix webhooks and remove the `Authorization` logic there _Closes #19872_ Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: Gusted <williamzijl7@hotmail.com> Co-authored-by: delvh <dev.lh@web.de>
2022-11-03 14:23:20 -04:00
}