TrueCloudLab/frostfs-node
20
1
Fork 27
Code Issues 100 Pull requests 4 Projects Releases 1 Activity Actions

object: Move target initialization to separate package #1344

Merged
fyrchik merged 1 commit from aarifullin/frostfs-node:feat/patch/refactor/1 into master 2024-09-05 13:04:00 +00:00
Conversation 0 Commits 1 Files changed 22 +599 -585
aarifullin commented 2024-08-30 09:10:13 +00:00
Member
Copy link
  • Split the logic of write target initialization to different packages;
  • Refactor patch and put services: since both service initialize the target
    themselves.

Close #1310

* Split the logic of write target initialization to different packages; * Refactor patch and put services: since both service initialize the target themselves. Close #1310
aarifullin added the
refactoring
 label 2024-08-30 09:10:13 +00:00
aarifullin added 1 commit 2024-08-30 09:10:19 +00:00
[#1310] object: Move target initialization to separate package
All checks were successful
Tests and linters / Run gofumpt (pull_request) Successful in 1m59s
Details
DCO action / DCO (pull_request) Successful in 2m1s
Details
Tests and linters / Tests (1.22) (pull_request) Successful in 3m45s
Details
Pre-commit hooks / Pre-commit (pull_request) Successful in 3m48s
Details
Tests and linters / Tests (1.23) (pull_request) Successful in 3m44s
Details
Vulncheck / Vulncheck (pull_request) Successful in 3m47s
Details
Tests and linters / Lint (pull_request) Successful in 4m3s
Details
Tests and linters / Staticcheck (pull_request) Successful in 4m54s
Details
Tests and linters / gopls check (pull_request) Successful in 5m16s
Details
Tests and linters / Tests with -race (pull_request) Successful in 5m23s
Details
Build / Build Components (1.22) (pull_request) Successful in 5m24s
Details
Build / Build Components (1.23) (pull_request) Successful in 5m23s
Details
3e1bd3a2f0 
* Split the logic of write target initialization to different packages;
* Refactor patch and put services: since both service initialize the target
  themselves.

Signed-off-by: Airat Arifullin <a.arifullin@yadro.com>
aarifullin force-pushed feat/patch/refactor/1 from 3e1bd3a2f0 to 1d42455131 2024-08-30 09:12:45 +00:00 Compare
aarifullin changed title from [#1310] object: Move target initialization to separate package to WIP: [#1310] object: Move target initialization to separate package 2024-08-30 09:13:02 +00:00
aarifullin force-pushed feat/patch/refactor/1 from 1d42455131 to 64ba2b0b4a 2024-08-30 09:17:06 +00:00 Compare
dstepanov-yadro approved these changes 2024-08-30 10:16:36 +00:00
Dismissed
aarifullin changed title from WIP: [#1310] object: Move target initialization to separate package to [#1310] object: Move target initialization to separate package 2024-09-02 07:27:15 +00:00
aarifullin requested review from dstepanov-yadro 2024-09-02 07:30:01 +00:00
fyrchik reviewed 2024-09-02 07:43:36 +00:00
pkg/services/object/common/target/target.go Outdated
@ -0,0 +17,4 @@
"git.frostfs.info/TrueCloudLab/frostfs-sdk-go/user"
)
type Params struct {
fyrchik commented 2024-09-02 07:43:35 +00:00
Owner
Copy link

Why is this field embedded, while the others are not?
IMO it is better to use explicit names in Params.

Why is this field embedded, while the others are not? IMO it is better to use explicit names in `Params`.
aarifullin commented 2024-09-02 10:03:37 +00:00
Author
Member
Copy link

Config is the field in Params. Please, check this out

`Config` is the field in `Params`. Please, check this out
fyrchik marked this conversation as resolved
pkg/services/object/common/target/target.go Outdated
@ -0,0 +32,4 @@
SignRequestPrivateKey *ecdsa.PrivateKey
}
fyrchik commented 2024-09-02 07:43:36 +00:00
Owner
Copy link

Why do we need a separate struct here, just to be able to convert it to the struct of the same shape?

Why do we need a separate struct here, just to be able to convert it to the struct of the same shape?
aarifullin commented 2024-09-02 09:00:06 +00:00
Author
Member
Copy link

Here are highlighted points that aren't related to specifically to this question, but which help to understand why packages were refactored in this way:

  1. The Target initialization logic should be moved in a common package. Since both patch and put can reuse this
  2. I assumed that writer and target should be distingushed:
  • writer package's purpose is to create transformer.ObjectWriter
  • writer package is used in a few packages
  • target package's purpose is to create transformer.ChunkedObjectWriter
  • target is used only within put and patch streams

It looked ambiguous with me to use objectwriter.Params as an input parameter for target initialization (New). I admit that such convertation looks ugly - so, I'll remove it

Here are highlighted points that aren't related to specifically to this question, but which help to understand why packages were refactored in this way: 1. The `Target` initialization logic should be moved in a common package. Since both `patch` and `put` can reuse this 2. I assumed that `writer` and `target` should be distingushed: - `writer` package's purpose is to create `transformer.ObjectWriter` - `writer` package is used in a few packages - `target` package's purpose is to create `transformer.ChunkedObjectWriter` - `target` is used only within `put` and `patch` streams It looked ambiguous with me to use `objectwriter.Params` as an input parameter for target initialization (`New`). I admit that such convertation looks ugly - so, I'll remove it
aarifullin commented 2024-09-02 10:04:27 +00:00
Author
Member
Copy link

Since target.New uses objectwriter.Params. Please, check this out

Since `target.New` uses `objectwriter.Params`. Please, check this out
fyrchik marked this conversation as resolved
pkg/services/object/common/writer/common.go Outdated
fyrchik commented 2024-09-02 07:43:36 +00:00
Owner
Copy link

Why is it public now?

Why is it public now?
aarifullin commented 2024-09-02 07:59:48 +00:00
Author
Member
Copy link

The file has been moved to another package (now it's within common/writer) and we need to access some fields

The file has been moved to another package (now it's within `common/writer`) and we need to access some fields
fyrchik marked this conversation as resolved
pkg/services/object/common/writer/writer.go Outdated
@ -0,0 +38,4 @@
type FormatValidatorConfig interface {
VerifySessionTokenIssuer() bool
}
fyrchik commented 2024-09-02 07:43:36 +00:00
Owner
Copy link

Why don't you use Config like in other code? "Static" is not even a noun (or doesn't have the meaning we are trying to convey here).

Why don't you use `Config` like in other code? "Static" is not even a noun (or doesn't have the meaning we are trying to convey here).
aarifullin commented 2024-09-02 08:39:54 +00:00
Author
Member
Copy link

Ok, I'll rename it back

Ok, I'll rename it back
aarifullin commented 2024-09-02 09:52:41 +00:00
Author
Member
Copy link

BTW, I have named it Static because I was not sure that using the word Config is appropriate for non-service package - as it was previosly in put service package.
Static - I meant that there parameters are set during initialzation only unlike field in Params

BTW, I have named it `Static` because I was not sure that using the word `Config` is appropriate for non-service package - as it was previosly in `put` service package. `Static` - I meant that there parameters are set during initialzation only unlike field in `Params`
fyrchik marked this conversation as resolved
pkg/services/object/put/single.go Outdated
@ -21,6 +21,7 @@ import (
"git.frostfs.info/TrueCloudLab/frostfs-node/pkg/core/object"
"git.frostfs.info/TrueCloudLab/frostfs-node/pkg/core/policy"
"git.frostfs.info/TrueCloudLab/frostfs-node/pkg/network"
fyrchik commented 2024-09-02 07:43:36 +00:00
Owner
Copy link

PutSingle seems to differ heavily from the regular PUT, as it uses multiple exported entities from our new packages, without using more coarse chunks.
Why is it so? It seems we send the same object to some target. Optimizations?

PutSingle seems to differ heavily from the regular PUT, as it uses multiple exported entities from our new packages, without using more coarse chunks. Why is it so? It seems we send the same object to some target. Optimizations?
aarifullin commented 2024-09-02 09:45:24 +00:00
Author
Member
Copy link

Ah... That's the hardest thing in this refactoring. The refactoring for single is not for optimization purpose. I believe that PutSingle is not related neither to writer nor to target. It still uses the config (currently it's Static that's going to be renamed) and NodeDescriptor.

Moving Static (Config) and common things like NodeDescriptor to one more separate package is not the big deal

Ah... That's the hardest thing in this refactoring. The refactoring for `single` is not for optimization purpose. I believe that `PutSingle` is not related neither to `writer` nor to `target`. It still uses the config (currently it's `Static` that's going to be renamed) and `NodeDescriptor`. Moving `Static` (`Config`) and common things like `NodeDescriptor` to one more separate package is not the big deal
fyrchik commented 2024-09-02 18:02:25 +00:00
Owner
Copy link

If we go one step further and move NodeIterator to a separate package (together with it's config), will it help somehow? Or is the issue deeper?

If we go one step further and move `NodeIterator` to a separate package (together with it's config), will it help somehow? Or is the issue deeper?
fyrchik marked this conversation as resolved
pkg/services/object/put/single.go Outdated
@ -37,2 +38,3 @@
var errInvalidPayloadChecksum = errors.New("incorrect payload checksum")
var (
errInvalidPayloadChecksum = errors.New("incorrect payload checksum")
fyrchik commented 2024-09-02 07:43:36 +00:00
Owner
Copy link

This error is still present in common/target/validation.go. Is it used there? Is is the same error as here?

This error is still present in `common/target/validation.go`. Is it used there? Is is the same error as here?
aarifullin commented 2024-09-02 10:13:38 +00:00
Author
Member
Copy link

These two errors have been moved to common/errors package as they used in both target and put packages. I would not like to import errors from package target because that one who will reading the codebase won't get the idea how should he has guessed to import these errors from the package target

These two errors have been moved to `common/errors` package as they used in both `target` and `put` packages. I would *not* like to import errors from package `target` because that one who will reading the codebase won't get the idea how should he has guessed to import these errors from the package `target`
fyrchik commented 2024-09-04 09:02:32 +00:00
Owner
Copy link

Frankly, I would avoid name clash with the package from stdlib.
Can we just reuse errors from target package? We are importing it anyway, so no new dependency occurs.

Frankly, I would avoid name clash with the package from `stdlib`. Can we just reuse errors from `target` package? We are importing it anyway, so no new dependency occurs.
fyrchik commented 2024-09-05 08:02:41 +00:00
Owner
Copy link

@aarifullin

@aarifullin
aarifullin commented 2024-09-05 08:10:07 +00:00
Author
Member
Copy link

Fixed

Fixed
pkg/services/object/put/streamer.go Outdated
@ -55,4 +53,4 @@
// MaxObjectSize returns maximum payload size for the streaming session.
//
// Must be called after the successful Init.
func (p *Streamer) MaxObjectSize() uint64 {
fyrchik commented 2024-09-02 07:43:36 +00:00
Owner
Copy link

Previously this method exported private field, what do we need this method for now?

Previously this method exported private field, what do we need this method for now?
aarifullin commented 2024-09-02 10:04:42 +00:00
Author
Member
Copy link

Removed

Removed
fyrchik marked this conversation as resolved
fyrchik changed title from [#1310] object: Move target initialization to separate package to object: Move target initialization to separate package 2024-09-02 07:43:53 +00:00
fyrchik requested review from storage-core-committers 2024-09-02 07:43:58 +00:00
fyrchik requested review from storage-core-developers 2024-09-02 07:43:59 +00:00
aarifullin force-pushed feat/patch/refactor/1 from 64ba2b0b4a to a85561bf25 2024-09-02 10:03:02 +00:00 Compare
aarifullin force-pushed feat/patch/refactor/1 from a85561bf25 to 28a1208240 2024-09-02 10:13:39 +00:00 Compare
aarifullin force-pushed feat/patch/refactor/1 from 28a1208240 to 90cf89b26c 2024-09-04 07:24:47 +00:00 Compare
dstepanov-yadro approved these changes 2024-09-04 07:27:21 +00:00
Dismissed
acid-ant approved these changes 2024-09-04 08:19:58 +00:00
Dismissed
aarifullin force-pushed feat/patch/refactor/1 from 90cf89b26c to 1fe5c82667 2024-09-05 08:09:44 +00:00 Compare
aarifullin dismissed dstepanov-yadro's review 2024-09-05 08:09:44 +00:00
Reason:

New commits pushed, approval review dismissed automatically according to repository settings

aarifullin dismissed dstepanov-yadro's review 2024-09-05 08:09:44 +00:00
Reason:

New commits pushed, approval review dismissed automatically according to repository settings

aarifullin dismissed acid-ant's review 2024-09-05 08:09:45 +00:00
Reason:

New commits pushed, approval review dismissed automatically according to repository settings

fyrchik reviewed 2024-09-05 08:30:49 +00:00
pkg/services/object/put/single.go Outdated
@ -21,6 +21,8 @@ import (
"git.frostfs.info/TrueCloudLab/frostfs-node/pkg/core/object"
"git.frostfs.info/TrueCloudLab/frostfs-node/pkg/core/policy"
"git.frostfs.info/TrueCloudLab/frostfs-node/pkg/network"
targeterrors "git.frostfs.info/TrueCloudLab/frostfs-node/pkg/services/object/common/target"
fyrchik commented 2024-09-05 08:30:49 +00:00
Owner
Copy link

Why use different alias here?

Why use different alias here?
aarifullin commented 2024-09-05 09:16:32 +00:00
Author
Member
Copy link

Oh, sorry. That was left after multiple renamings.
I have removed all aliases for common/target package because this package's name seems unambiguous

Oh, sorry. That was left after multiple renamings. I have removed all aliases for `common/target` package because this package's name seems unambiguous
aarifullin force-pushed feat/patch/refactor/1 from 1fe5c82667 to 18220985bf 2024-09-05 09:14:37 +00:00 Compare
fyrchik merged commit b3deb893ba into master 2024-09-05 13:04:00 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
dstepanov-yadro
TrueCloudLab/storage-core-developers
Labels
Clear labels   
P0

Highest

  
P1

High

  
P2

Medium

  
P3

Low

  
badger

Badger DB related issues

  
frostfs-adm

   
frostfs-cli

   
frostfs-ir

   
frostfs-lens

   
frostfs-node

   
good first issue

Good for newcomers

  
triage
  
Infrastructure

Infrastructure, Automation, CI/CD or DevOps related stuff

  
blocked

The issue or PR is blocked by something

  
bug

Something is not working

  
config

Configuration format update or breaking change

  
discussion

Talking about something in order to reach a decision or to exchange ideas

  
documentation

Documentation related

  
duplicate

This issue or pull request already exists

  
enhancement

New feature

  
go

Language features adoption

  
help wanted

Extra attention is needed

  
internal

   
invalid

Something is wrong

  
kludge

This is a kludge and we know it

  
observability

Related to metrics, tracing and logs

  
perfomance

Perfomance related

  
question

More information is needed

  
refactoring

Refactoring

  
wontfix

This won't be fixed

No labels
P0
P1
P2
P3
badger
frostfs-adm
frostfs-cli
frostfs-ir
frostfs-lens
frostfs-node
good first issue
triage
Infrastructure
blocked
bug
config
discussion
documentation
duplicate
enhancement
go
help wanted
internal
invalid
kludge
observability
perfomance
question
refactoring
wontfix
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
4 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: TrueCloudLab/frostfs-node#1344
No description provided.
Delete branch "aarifullin/frostfs-node:feat/patch/refactor/1"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?