0
0
mirror of https://github.com/go-gitea/gitea.git synced 2024-12-22 16:36:04 +01:00
gitea/modules/git/foreachref/format.go

84 lines
2.5 KiB
Go
Raw Normal View History

Improve sync performance for pull-mirrors (#19125) This addresses https://github.com/go-gitea/gitea/issues/18352 It aims to improve performance (and resource use) of the `SyncReleasesWithTags` operation for pull-mirrors. For large repositories with many tags, `SyncReleasesWithTags` can be a costly operation (taking several minutes to complete). The reason is two-fold: 1. on sync, every upstream repo tag is compared (for changes) against existing local entries in the release table to ensure that they are up-to-date. 2. the procedure for getting _each tag_ involves a series of git operations ```bash git show-ref --tags -- v8.2.4477 git cat-file -t 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git cat-file -p 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git rev-list --count 29ab6ce9f36660cffaad3c8789e71162e5db5d2f ``` of which the `git rev-list --count` can be particularly heavy. This PR optimizes performance for pull-mirrors. We utilize the fact that a pull-mirror is always identical to its upstream and rebuild the entire release table on every sync and use a batch `git for-each-ref .. refs/tags` call to retrieve all tags in one go. For large mirror repos, with hundreds of annotated tags, this brings down the duration of the sync operation from several minutes to a few seconds. A few unscientific examples run on my local machine: - https://github.com/spring-projects/spring-boot (223 tags) - before: `0m28,673s` - after: `0m2,244s` - https://github.com/kubernetes/kubernetes (890 tags) - before: `8m00s` - after: `0m8,520s` - https://github.com/vim/vim (13954 tags) - before: `14m20,383s` - after: `0m35,467s` I added a `foreachref` package which contains a flexible way of specifying which reference fields are of interest (`git-for-each-ref(1)`) and to produce a parser for the expected output. These could be reused in other places where `for-each-ref` is used. I'll add unit tests for those if the overall PR looks promising.
2022-03-31 14:30:40 +02:00
// Copyright 2022 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
Improve sync performance for pull-mirrors (#19125) This addresses https://github.com/go-gitea/gitea/issues/18352 It aims to improve performance (and resource use) of the `SyncReleasesWithTags` operation for pull-mirrors. For large repositories with many tags, `SyncReleasesWithTags` can be a costly operation (taking several minutes to complete). The reason is two-fold: 1. on sync, every upstream repo tag is compared (for changes) against existing local entries in the release table to ensure that they are up-to-date. 2. the procedure for getting _each tag_ involves a series of git operations ```bash git show-ref --tags -- v8.2.4477 git cat-file -t 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git cat-file -p 29ab6ce9f36660cffaad3c8789e71162e5db5d2f git rev-list --count 29ab6ce9f36660cffaad3c8789e71162e5db5d2f ``` of which the `git rev-list --count` can be particularly heavy. This PR optimizes performance for pull-mirrors. We utilize the fact that a pull-mirror is always identical to its upstream and rebuild the entire release table on every sync and use a batch `git for-each-ref .. refs/tags` call to retrieve all tags in one go. For large mirror repos, with hundreds of annotated tags, this brings down the duration of the sync operation from several minutes to a few seconds. A few unscientific examples run on my local machine: - https://github.com/spring-projects/spring-boot (223 tags) - before: `0m28,673s` - after: `0m2,244s` - https://github.com/kubernetes/kubernetes (890 tags) - before: `8m00s` - after: `0m8,520s` - https://github.com/vim/vim (13954 tags) - before: `14m20,383s` - after: `0m35,467s` I added a `foreachref` package which contains a flexible way of specifying which reference fields are of interest (`git-for-each-ref(1)`) and to produce a parser for the expected output. These could be reused in other places where `for-each-ref` is used. I'll add unit tests for those if the overall PR looks promising.
2022-03-31 14:30:40 +02:00
package foreachref
import (
"encoding/hex"
"fmt"
"io"
"strings"
)
var (
nullChar = []byte("\x00")
dualNullChar = []byte("\x00\x00")
)
// Format supports specifying and parsing an output format for 'git
// for-each-ref'. See See git-for-each-ref(1) for available fields.
type Format struct {
// fieldNames hold %(fieldname)s to be passed to the '--format' flag of
// for-each-ref. See git-for-each-ref(1) for available fields.
fieldNames []string
// fieldDelim is the character sequence that is used to separate fields
// for each reference. fieldDelim and refDelim should be selected to not
// interfere with each other and to not be present in field values.
fieldDelim []byte
// fieldDelimStr is a string representation of fieldDelim. Used to save
// us from repetitive reallocation whenever we need the delimiter as a
// string.
fieldDelimStr string
// refDelim is the character sequence used to separate reference from
// each other in the output. fieldDelim and refDelim should be selected
// to not interfere with each other and to not be present in field
// values.
refDelim []byte
}
// NewFormat creates a forEachRefFormat using the specified fieldNames. See
// git-for-each-ref(1) for available fields.
func NewFormat(fieldNames ...string) Format {
return Format{
fieldNames: fieldNames,
fieldDelim: nullChar,
fieldDelimStr: string(nullChar),
refDelim: dualNullChar,
}
}
// Flag returns a for-each-ref --format flag value that captures the fieldNames.
func (f Format) Flag() string {
var formatFlag strings.Builder
for i, field := range f.fieldNames {
// field key and field value
formatFlag.WriteString(fmt.Sprintf("%s %%(%s)", field, field))
if i < len(f.fieldNames)-1 {
// note: escape delimiters to allow control characters as
// delimiters. For example, '%00' for null character or '%0a'
// for newline.
formatFlag.WriteString(f.hexEscaped(f.fieldDelim))
}
}
formatFlag.WriteString(f.hexEscaped(f.refDelim))
return formatFlag.String()
}
// Parser returns a Parser capable of parsing 'git for-each-ref' output produced
// with this Format.
func (f Format) Parser(r io.Reader) *Parser {
return NewParser(r, f)
}
// hexEscaped produces hex-escpaed characters from a string. For example, "\n\0"
// would turn into "%0a%00".
func (f Format) hexEscaped(delim []byte) string {
escaped := ""
for i := 0; i < len(delim); i++ {
escaped += "%" + hex.EncodeToString([]byte{delim[i]})
}
return escaped
}