Archive Module

Functions for creating and extracting archive files.

archive.extract

Extracts contents from an archive file.

archive.extract(
    archive = ctx.path("deps/library.tar.gz"),
    type = "AUTO",
    destination_folder = ctx.path("vendor/library"),
    paths = glob(["src/**"]),
)

Parameters

Parameter	Type	Description
archive	path	Path to the archive file
type	string	Archive type: AUTO, JAR, ZIP, TAR, TAR_GZ, TAR_XZ, TAR_BZ2
destination_folder	path	Where to extract (defaults to archive’s directory)
paths	glob	Optional filter for which files to extract

Supported Formats

Extension	Type
.zip	ZIP
.jar	JAR
.tar	TAR
.tgz, .tar.gz	TAR_GZ
.tar.xz	TAR_XZ
.tar.bz2	TAR_BZ2

archive.create

Creates an archive from files in the working directory.

archive.create(
    archive = ctx.path("output/release.tar.gz"),
    files = glob(["dist/**", "README.md"]),
)

Parameters

Parameter	Type	Description
archive	path	Output path for the archive
files	glob	Files to include (defaults to all files)

The archive type is inferred from the file extension.

Use Cases

Vendor Dependencies

Extract a third-party library during transformation:

def _extract_deps(ctx):
    archive.extract(
        archive = ctx.path("third_party/protobuf.tar.gz"),
        destination_folder = ctx.path("third_party/protobuf"),
    )
    # Remove the archive after extraction
    ctx.remove(glob(["third_party/*.tar.gz"]))

Package Artifacts

Create a distributable archive:

def _create_release(ctx):
    archive.create(
        archive = ctx.path("release.zip"),
        files = glob([
            "bin/**",
            "lib/**",
            "LICENSE",
            "README.md",
        ]),
    )

Selective Extraction

Extract only specific files from a large archive:

archive.extract(
    archive = ctx.path("sdk.zip"),
    paths = glob(["headers/**/*.h", "lib/*.a"]),
    destination_folder = ctx.path("vendor/sdk"),
)

Notes

• The AUTO type detection relies on file extensions
• Original file paths are preserved when creating archives
• Extraction creates directories as needed