Skip to content

experimental/air: parallel gzip for the plain_tar snapshot packer #6571

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
ben-hansen-db wants to merge 3 commits into
base: main
Choose a base branch
from
+35 −13
Open

experimental/air: parallel gzip for the plain_tar snapshot packer #6571

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
fbe71c9
experimental/air: parallel gzip for the plain_tar snapshot packer
ben-hansen-db Sep 8, 2026
e3b164d
experimental/air: use default gzip level instead of BestSpeed
ben-hansen-db Sep 8, 2026
8155ccf
experimental/air: trim the pgzip comment per review
ben-hansen-db Sep 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions NOTICE
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,6 +131,10 @@ jackc/pgx - https://github.com/jackc/pgx
Copyright (c) 2013-2021 Jack Christensen
License - https://github.com/jackc/pgx/blob/master/LICENSE

klauspost/pgzip - https://github.com/klauspost/pgzip
Copyright (c) 2014 Klaus Post
License - https://github.com/klauspost/pgzip/blob/master/LICENSE

charmbracelet/bubbles - https://github.com/charmbracelet/bubbles
Copyright (c) 2020-2025 Charmbracelet, Inc
License - https://github.com/charmbracelet/bubbles/blob/master/LICENSE
Expand Down
38 changes: 25 additions & 13 deletions experimental/air/cmd/snapshot_package.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,8 @@ import (
"os/exec"
"path/filepath"
"strings"

@vinchenzo-db vinchenzo-db Sep 8, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is Go semantics/lint to have this empty space? if not then rm

@ben-hansen-db ben-hansen-db Sep 8, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

need to keep it according to claude:

that blank line is the standard import grouping (stdlib vs third-party) that goimports/gofumpt enforce, and this repo runs both as formatters. Without it, gofmt sorts github.com/klauspost/pgzip alphabetically into the stdlib block (between fmt and os) and goimports re-adds the separator, so ./task fmt would put it right back.

"github.com/klauspost/pgzip"
)

// Tar builders ported from cli/utils/snapshot.py. Both shell out (git archive / tar)
Expand Down Expand Up @@ -41,47 +43,57 @@ func createGitArchiveSnapshot(ctx context.Context, git gitRepo, commitSHA, outpu
// excluded; a .gitignore at repoPath is honored.
func createPlainTarball(ctx context.Context, repoPath, outputTarball string, includePaths []string, isGitRepo bool) error {
dirName := filepath.Base(repoPath)
// Absolute so it resolves correctly regardless of tar's working dir (set below).
// Absolute so it resolves correctly regardless of tar's working dir.
parent, err := filepath.Abs(filepath.Dir(repoPath))
if err != nil {
return err
}

// Pass the archive path relative to its own directory (run tar there), never a
// full path: on Windows an absolute path like `C:\out\x.tar.gz` makes tar read
// the `C:` as a remote host ("Cannot connect to C:"), since tar treats a colon
// in the -f arg as host:path. A bare basename with -C avoids that on GNU tar and
// bsdtar alike.
outDirAbs, err := filepath.Abs(filepath.Dir(outputTarball))
files, err := snapshotFiles(ctx, repoPath, includePaths, isGitRepo)
if err != nil {
return err
}
outName := filepath.Base(outputTarball)

files, err := snapshotFiles(ctx, repoPath, includePaths, isGitRepo)
out, err := os.Create(outputTarball)
if err != nil {
return fmt.Errorf("failed to create tarball: %w", err)
}

// tar writes the uncompressed archive to stdout and we gzip it with klauspost/pgzip
// rather than tar's single-threaded -z, which dominates packaging time on a large
// tree (pgzip spreads the same compression across cores). Compressing outside tar
// also keeps any archive path out of tar's args, avoiding the Windows colon-in-path
// issue the `-f <path>` form otherwise hits.
gz, err := pgzip.NewWriterLevel(out, pgzip.DefaultCompression)
if err != nil {
out.Close()
return err
}
args := []string{"-czf", outName, "-C", parent, "--null", "--no-recursion", "-T", "-"}

args := []string{"-cf", "-", "-C", parent, "--null", "--no-recursion", "-T", "-"}
cmd := exec.CommandContext(ctx, "tar", args...)
// Run tar in the output directory so the bare -f basename lands there.
cmd.Dir = outDirAbs
var stdin bytes.Buffer
for _, file := range files {
stdin.WriteString(filepath.ToSlash(filepath.Join(dirName, file)))
stdin.WriteByte(0)
}
cmd.Stdin = &stdin
cmd.Stdout = gz
var stderr bytes.Buffer
cmd.Stderr = &stderr
if err := cmd.Run(); err != nil {
gz.Close()
out.Close()
if msg := strings.TrimSpace(stderr.String()); msg != "" {
return fmt.Errorf("failed to create plain tarball: %w: %s", err, msg)
}
return fmt.Errorf("failed to create plain tarball: %w", err)
}
return nil
if err := gz.Close(); err != nil {
out.Close()
return fmt.Errorf("failed to finalize gzip: %w", err)
}
return out.Close()
}

func snapshotFiles(ctx context.Context, repoPath string, includePaths []string, isGitRepo bool) ([]string, error) {
Expand Down
2 changes: 2 additions & 0 deletions go.mod
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,7 @@ require (
github.com/hashicorp/terraform-json v0.28.0 // MPL-2.0
github.com/hexops/gotextdiff v1.0.3 // BSD-3-Clause
github.com/jackc/pgx/v5 v5.10.0 // MIT
github.com/klauspost/pgzip v1.2.6 // MIT
github.com/mattn/go-isatty v0.0.24 // MIT
github.com/muesli/termenv v0.16.0 // MIT
github.com/palantir/pkg/yamlpatch v1.5.0 // BSD-3-Clause
Expand Down Expand Up @@ -87,6 +88,7 @@ require (
github.com/inconshreveable/mousetrap v1.1.0 // indirect
github.com/jackc/pgpassfile v1.0.0 // indirect
github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 // indirect
github.com/klauspost/compress v1.19.2 // indirect
github.com/lucasb-eyer/go-colorful v1.4.0 // indirect
github.com/mattn/go-colorable v0.1.14 // indirect
github.com/mattn/go-localereader v0.0.1 // indirect
Expand Down
4 changes: 4 additions & 0 deletions go.sum
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -156,8 +156,12 @@ github.com/jbenet/go-context v0.0.0-20150711004518-d14ea06fba99 h1:BQSFePA1RWJOl
github.com/jbenet/go-context v0.0.0-20150711004518-d14ea06fba99/go.mod h1:1lJo3i6rXxKeerYnT8Nvf0QmHCRC1n8sfWVwXF2Frvo=
github.com/kevinburke/ssh_config v1.2.0 h1:x584FjTGwHzMwvHx18PXxbBVzfnxogHaAReU4gf13a4=
github.com/kevinburke/ssh_config v1.2.0/go.mod h1:CT57kijsi8u/K/BOFA39wgDQJ9CxiF4nAY/ojJ6r6mM=
github.com/klauspost/compress v1.19.2 h1:hMRETovs/pu/dVWN7zIT1PGG8t509MwT6bO7XSi26R8=
github.com/klauspost/compress v1.19.2/go.mod h1:cwPg85FWrGar70rWktvGQj8/hthj3wpl0PGDogxkrSQ=
github.com/klauspost/cpuid/v2 v2.3.0 h1:S4CRMLnYUhGeDFDqkGriYKdfoFlDnMtqTiI/sFzhA9Y=
github.com/klauspost/cpuid/v2 v2.3.0/go.mod h1:hqwkgyIinND0mEev00jJYCxPNVRVXFQeu1XKlok6oO0=
github.com/klauspost/pgzip v1.2.6 h1:8RXeL5crjEUFnR2/Sn6GJNWtSQ3Dk8pq4CL3jvdDyjU=
github.com/klauspost/pgzip v1.2.6/go.mod h1:Ch1tH69qFZu15pkjo5kYi6mth2Zzwzt50oCQKQE9RUs=
github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
Expand Down
Loading