forgejo/modules/git/repo_branch_nogogit.go
wxiaoguang 6bc3079c00
Refactor git command package to improve security and maintainability (#22678)
This PR follows #21535 (and replace #22592)

## Review without space diff

https://github.com/go-gitea/gitea/pull/22678/files?diff=split&w=1

## Purpose of this PR

1. Make git module command completely safe (risky user inputs won't be
passed as argument option anymore)
2. Avoid low-level mistakes like
https://github.com/go-gitea/gitea/pull/22098#discussion_r1045234918
3. Remove deprecated and dirty `CmdArgCheck` function, hide the `CmdArg`
type
4. Simplify code when using git command

## The main idea of this PR

* Move the `git.CmdArg` to the `internal` package, then no other package
except `git` could use it. Then developers could never do
`AddArguments(git.CmdArg(userInput))` any more.
* Introduce `git.ToTrustedCmdArgs`, it's for user-provided and already
trusted arguments. It's only used in a few cases, for example: use git
arguments from config file, help unit test with some arguments.
* Introduce `AddOptionValues` and `AddOptionFormat`, they make code more
clear and simple:
    * Before: `AddArguments("-m").AddDynamicArguments(message)`
    * After: `AddOptionValues("-m", message)`
    * -
* Before: `AddArguments(git.CmdArg(fmt.Sprintf("--author='%s <%s>'",
sig.Name, sig.Email)))`
* After: `AddOptionFormat("--author='%s <%s>'", sig.Name, sig.Email)`

## FAQ

### Why these changes were not done in #21535 ?

#21535 is mainly a search&replace, it did its best to not change too
much logic.

Making the framework better needs a lot of changes, so this separate PR
is needed as the second step.


### The naming of `AddOptionXxx`

According to git's manual, the `--xxx` part is called `option`.

### How can it guarantee that `internal.CmdArg` won't be not misused?

Go's specification guarantees that. Trying to access other package's
internal package causes compilation error.

And, `golangci-lint` also denies the git/internal package. Only the
`git/command.go` can use it carefully.

### There is still a `ToTrustedCmdArgs`, will it still allow developers
to make mistakes and pass untrusted arguments?

Generally speaking, no. Because when using `ToTrustedCmdArgs`, the code
will be very complex (see the changes for examples). Then developers and
reviewers can know that something might be unreasonable.

### Why there was a `CmdArgCheck` and why it's removed?

At the moment of #21535, to reduce unnecessary changes, `CmdArgCheck`
was introduced as a hacky patch. Now, almost all code could be written
as `cmd := NewCommand(); cmd.AddXxx(...)`, then there is no need for
`CmdArgCheck` anymore.


### Why many codes for `signArg == ""` is deleted?

Because in the old code, `signArg` could never be empty string, it's
either `-S[key-id]` or `--no-gpg-sign`. So the `signArg == ""` is just
dead code.

---------

Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
2023-02-04 10:30:43 +08:00

199 lines
5.3 KiB
Go

// Copyright 2015 The Gogs Authors. All rights reserved.
// Copyright 2018 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
//go:build !gogit
package git
import (
"bufio"
"bytes"
"context"
"io"
"strings"
"code.gitea.io/gitea/modules/log"
)
// IsObjectExist returns true if given reference exists in the repository.
func (repo *Repository) IsObjectExist(name string) bool {
if name == "" {
return false
}
wr, rd, cancel := repo.CatFileBatchCheck(repo.Ctx)
defer cancel()
_, err := wr.Write([]byte(name + "\n"))
if err != nil {
log.Debug("Error writing to CatFileBatchCheck %v", err)
return false
}
sha, _, _, err := ReadBatchLine(rd)
return err == nil && bytes.HasPrefix(sha, []byte(strings.TrimSpace(name)))
}
// IsReferenceExist returns true if given reference exists in the repository.
func (repo *Repository) IsReferenceExist(name string) bool {
if name == "" {
return false
}
wr, rd, cancel := repo.CatFileBatchCheck(repo.Ctx)
defer cancel()
_, err := wr.Write([]byte(name + "\n"))
if err != nil {
log.Debug("Error writing to CatFileBatchCheck %v", err)
return false
}
_, _, _, err = ReadBatchLine(rd)
return err == nil
}
// IsBranchExist returns true if given branch exists in current repository.
func (repo *Repository) IsBranchExist(name string) bool {
if repo == nil || name == "" {
return false
}
return repo.IsReferenceExist(BranchPrefix + name)
}
// GetBranchNames returns branches from the repository, skipping "skip" initial branches and
// returning at most "limit" branches, or all branches if "limit" is 0.
func (repo *Repository) GetBranchNames(skip, limit int) ([]string, int, error) {
return callShowRef(repo.Ctx, repo.Path, BranchPrefix, TrustedCmdArgs{BranchPrefix, "--sort=-committerdate"}, skip, limit)
}
// WalkReferences walks all the references from the repository
func WalkReferences(ctx context.Context, repoPath string, walkfn func(sha1, refname string) error) (int, error) {
return walkShowRef(ctx, repoPath, nil, 0, 0, walkfn)
}
// WalkReferences walks all the references from the repository
// refType should be empty, ObjectTag or ObjectBranch. All other values are equivalent to empty.
func (repo *Repository) WalkReferences(refType ObjectType, skip, limit int, walkfn func(sha1, refname string) error) (int, error) {
var args TrustedCmdArgs
switch refType {
case ObjectTag:
args = TrustedCmdArgs{TagPrefix, "--sort=-taggerdate"}
case ObjectBranch:
args = TrustedCmdArgs{BranchPrefix, "--sort=-committerdate"}
}
return walkShowRef(repo.Ctx, repo.Path, args, skip, limit, walkfn)
}
// callShowRef return refs, if limit = 0 it will not limit
func callShowRef(ctx context.Context, repoPath, trimPrefix string, extraArgs TrustedCmdArgs, skip, limit int) (branchNames []string, countAll int, err error) {
countAll, err = walkShowRef(ctx, repoPath, extraArgs, skip, limit, func(_, branchName string) error {
branchName = strings.TrimPrefix(branchName, trimPrefix)
branchNames = append(branchNames, branchName)
return nil
})
return branchNames, countAll, err
}
func walkShowRef(ctx context.Context, repoPath string, extraArgs TrustedCmdArgs, skip, limit int, walkfn func(sha1, refname string) error) (countAll int, err error) {
stdoutReader, stdoutWriter := io.Pipe()
defer func() {
_ = stdoutReader.Close()
_ = stdoutWriter.Close()
}()
go func() {
stderrBuilder := &strings.Builder{}
args := TrustedCmdArgs{"for-each-ref", "--format=%(objectname) %(refname)"}
args = append(args, extraArgs...)
err := NewCommand(ctx, args...).Run(&RunOpts{
Dir: repoPath,
Stdout: stdoutWriter,
Stderr: stderrBuilder,
})
if err != nil {
if stderrBuilder.Len() == 0 {
_ = stdoutWriter.Close()
return
}
_ = stdoutWriter.CloseWithError(ConcatenateError(err, stderrBuilder.String()))
} else {
_ = stdoutWriter.Close()
}
}()
i := 0
bufReader := bufio.NewReader(stdoutReader)
for i < skip {
_, isPrefix, err := bufReader.ReadLine()
if err == io.EOF {
return i, nil
}
if err != nil {
return 0, err
}
if !isPrefix {
i++
}
}
for limit == 0 || i < skip+limit {
// The output of show-ref is simply a list:
// <sha> SP <ref> LF
sha, err := bufReader.ReadString(' ')
if err == io.EOF {
return i, nil
}
if err != nil {
return 0, err
}
branchName, err := bufReader.ReadString('\n')
if err == io.EOF {
// This shouldn't happen... but we'll tolerate it for the sake of peace
return i, nil
}
if err != nil {
return i, err
}
if len(branchName) > 0 {
branchName = branchName[:len(branchName)-1]
}
if len(sha) > 0 {
sha = sha[:len(sha)-1]
}
err = walkfn(sha, branchName)
if err != nil {
return i, err
}
i++
}
// count all refs
for limit != 0 {
_, isPrefix, err := bufReader.ReadLine()
if err == io.EOF {
return i, nil
}
if err != nil {
return 0, err
}
if !isPrefix {
i++
}
}
return i, nil
}
// GetRefsBySha returns all references filtered with prefix that belong to a sha commit hash
func (repo *Repository) GetRefsBySha(sha, prefix string) ([]string, error) {
var revList []string
_, err := walkShowRef(repo.Ctx, repo.Path, nil, 0, 0, func(walkSha, refname string) error {
if walkSha == sha && strings.HasPrefix(refname, prefix) {
revList = append(revList, refname)
}
return nil
})
return revList, err
}